Xuban commited on
Commit
c817289
·
verified ·
1 Parent(s): 7d8fe07

Add reproducible LibreEdgeTAM snapshot

Browse files
LICENSE ADDED
@@ -0,0 +1,201 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Apache License
2
+ Version 2.0, January 2004
3
+ http://www.apache.org/licenses/
4
+
5
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
6
+
7
+ 1. Definitions.
8
+
9
+ "License" shall mean the terms and conditions for use, reproduction,
10
+ and distribution as defined by Sections 1 through 9 of this document.
11
+
12
+ "Licensor" shall mean the copyright owner or entity authorized by
13
+ the copyright owner that is granting the License.
14
+
15
+ "Legal Entity" shall mean the union of the acting entity and all
16
+ other entities that control, are controlled by, or are under common
17
+ control with that entity. For the purposes of this definition,
18
+ "control" means (i) the power, direct or indirect, to cause the
19
+ direction or management of such entity, whether by contract or
20
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
21
+ outstanding shares, or (iii) beneficial ownership of such entity.
22
+
23
+ "You" (or "Your") shall mean an individual or Legal Entity
24
+ exercising permissions granted by this License.
25
+
26
+ "Source" form shall mean the preferred form for making modifications,
27
+ including but not limited to software source code, documentation
28
+ source, and configuration files.
29
+
30
+ "Object" form shall mean any form resulting from mechanical
31
+ transformation or translation of a Source form, including but
32
+ not limited to compiled object code, generated documentation,
33
+ and conversions to other media types.
34
+
35
+ "Work" shall mean the work of authorship, whether in Source or
36
+ Object form, made available under the License, as indicated by a
37
+ copyright notice that is included in or attached to the work
38
+ (an example is provided in the Appendix below).
39
+
40
+ "Derivative Works" shall mean any work, whether in Source or Object
41
+ form, that is based on (or derived from) the Work and for which the
42
+ editorial revisions, annotations, elaborations, or other modifications
43
+ represent, as a whole, an original work of authorship. For the purposes
44
+ of this License, Derivative Works shall not include works that remain
45
+ separable from, or merely link (or bind by name) to the interfaces of,
46
+ the Work and Derivative Works thereof.
47
+
48
+ "Contribution" shall mean any work of authorship, including
49
+ the original version of the Work and any modifications or additions
50
+ to that Work or Derivative Works thereof, that is intentionally
51
+ submitted to Licensor for inclusion in the Work by the copyright owner
52
+ or by an individual or Legal Entity authorized to submit on behalf of
53
+ the copyright owner. For the purposes of this definition, "submitted"
54
+ means any form of electronic, verbal, or written communication sent
55
+ to the Licensor or its representatives, including but not limited to
56
+ communication on electronic mailing lists, source code control systems,
57
+ and issue tracking systems that are managed by, or on behalf of, the
58
+ Licensor for the purpose of discussing and improving the Work, but
59
+ excluding communication that is conspicuously marked or otherwise
60
+ designated in writing by the copyright owner as "Not a Contribution."
61
+
62
+ "Contributor" shall mean Licensor and any individual or Legal Entity
63
+ on behalf of whom a Contribution has been received by Licensor and
64
+ subsequently incorporated within the Work.
65
+
66
+ 2. Grant of Copyright License. Subject to the terms and conditions of
67
+ this License, each Contributor hereby grants to You a perpetual,
68
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
69
+ copyright license to reproduce, prepare Derivative Works of,
70
+ publicly display, publicly perform, sublicense, and distribute the
71
+ Work and such Derivative Works in Source or Object form.
72
+
73
+ 3. Grant of Patent License. Subject to the terms and conditions of
74
+ this License, each Contributor hereby grants to You a perpetual,
75
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
76
+ (except as stated in this section) patent license to make, have made,
77
+ use, offer to sell, sell, import, and otherwise transfer the Work,
78
+ where such license applies only to those patent claims licensable
79
+ by such Contributor that are necessarily infringed by their
80
+ Contribution(s) alone or by combination of their Contribution(s)
81
+ with the Work to which such Contribution(s) was submitted. If You
82
+ institute patent litigation against any entity (including a
83
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
84
+ or a Contribution incorporated within the Work constitutes direct
85
+ or contributory patent infringement, then any patent licenses
86
+ granted to You under this License for that Work shall terminate
87
+ as of the date such litigation is filed.
88
+
89
+ 4. Redistribution. You may reproduce and distribute copies of the
90
+ Work or Derivative Works thereof in any medium, with or without
91
+ modifications, and in Source or Object form, provided that You
92
+ meet the following conditions:
93
+
94
+ (a) You must give any other recipients of the Work or
95
+ Derivative Works a copy of this License; and
96
+
97
+ (b) You must cause any modified files to carry prominent notices
98
+ stating that You changed the files; and
99
+
100
+ (c) You must retain, in the Source form of any Derivative Works
101
+ that You distribute, all copyright, patent, trademark, and
102
+ attribution notices from the Source form of the Work,
103
+ excluding those notices that do not pertain to any part of
104
+ the Derivative Works; and
105
+
106
+ (d) If the Work includes a "NOTICE" text file as part of its
107
+ distribution, then any Derivative Works that You distribute must
108
+ include a readable copy of the attribution notices contained
109
+ within such NOTICE file, excluding those notices that do not
110
+ pertain to any part of the Derivative Works, in at least one
111
+ of the following places: within a NOTICE text file distributed
112
+ as part of the Derivative Works; within the Source form or
113
+ documentation, if provided along with the Derivative Works; or,
114
+ within a display generated by the Derivative Works, if and
115
+ wherever such third-party notices normally appear. The contents
116
+ of the NOTICE file are for informational purposes only and
117
+ do not modify the License. You may add Your own attribution
118
+ notices within Derivative Works that You distribute, alongside
119
+ or as an addendum to the NOTICE text from the Work, provided
120
+ that such additional attribution notices cannot be construed
121
+ as modifying the License.
122
+
123
+ You may add Your own copyright statement to Your modifications and
124
+ may provide additional or different license terms and conditions
125
+ for use, reproduction, or distribution of Your modifications, or
126
+ for any such Derivative Works as a whole, provided Your use,
127
+ reproduction, and distribution of the Work otherwise complies with
128
+ the conditions stated in this License.
129
+
130
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
131
+ any Contribution intentionally submitted for inclusion in the Work
132
+ by You to the Licensor shall be under the terms and conditions of
133
+ this License, without any additional terms or conditions.
134
+ Notwithstanding the above, nothing herein shall supersede or modify
135
+ the terms of any separate license agreement you may have executed
136
+ with Licensor regarding such Contributions.
137
+
138
+ 6. Trademarks. This License does not grant permission to use the trade
139
+ names, trademarks, service marks, or product names of the Licensor,
140
+ except as required for reasonable and customary use in describing the
141
+ origin of the Work and reproducing the content of the NOTICE file.
142
+
143
+ 7. Disclaimer of Warranty. Unless required by applicable law or
144
+ agreed to in writing, Licensor provides the Work (and each
145
+ Contributor provides its Contributions) on an "AS IS" BASIS,
146
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
147
+ implied, including, without limitation, any warranties or conditions
148
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
149
+ PARTICULAR PURPOSE. You are solely responsible for determining the
150
+ appropriateness of using or redistributing the Work and assume any
151
+ risks associated with Your exercise of permissions under this License.
152
+
153
+ 8. Limitation of Liability. In no event and under no legal theory,
154
+ whether in tort (including negligence), contract, or otherwise,
155
+ unless required by applicable law (such as deliberate and grossly
156
+ negligent acts) or agreed to in writing, shall any Contributor be
157
+ liable to You for damages, including any direct, indirect, special,
158
+ incidental, or consequential damages of any character arising as a
159
+ result of this License or out of the use or inability to use the
160
+ Work (including but not limited to damages for loss of goodwill,
161
+ work stoppage, computer failure or malfunction, or any and all
162
+ other commercial damages or losses), even if such Contributor
163
+ has been advised of the possibility of such damages.
164
+
165
+ 9. Accepting Warranty or Additional Liability. While redistributing
166
+ the Work or Derivative Works thereof, You may choose to offer,
167
+ and charge a fee for, acceptance of support, warranty, indemnity,
168
+ or other liability obligations and/or rights consistent with this
169
+ License. However, in accepting such obligations, You may act only
170
+ on Your own behalf and on Your sole responsibility, not on behalf
171
+ of any other Contributor, and only if You agree to indemnify,
172
+ defend, and hold each Contributor harmless for any liability
173
+ incurred by, or claims asserted against, such Contributor by reason
174
+ of your accepting any such warranty or additional liability.
175
+
176
+ END OF TERMS AND CONDITIONS
177
+
178
+ APPENDIX: How to apply the Apache License to your work.
179
+
180
+ To apply the Apache License to your work, attach the following
181
+ boilerplate notice, with the fields enclosed by brackets "[]"
182
+ replaced with your own identifying information. (Don't include
183
+ the brackets!) The text should be enclosed in the appropriate
184
+ comment syntax for the file format. We also recommend that a
185
+ file or class name and description of purpose be included on the
186
+ same "printed page" as the copyright notice for easier
187
+ identification within third-party archives.
188
+
189
+ Copyright [yyyy] [name of copyright owner]
190
+
191
+ Licensed under the Apache License, Version 2.0 (the "License");
192
+ you may not use this file except in compliance with the License.
193
+ You may obtain a copy of the License at
194
+
195
+ http://www.apache.org/licenses/LICENSE-2.0
196
+
197
+ Unless required by applicable law or agreed to in writing, software
198
+ distributed under the License is distributed on an "AS IS" BASIS,
199
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
200
+ See the License for the specific language governing permissions and
201
+ limitations under the License.
NOTICE ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ LibreEdgeTAM provenance notice
2
+
3
+ Official model
4
+ --------------
5
+ Repository: https://huggingface.co/facebook/EdgeTAM
6
+ Revision: 14d7ecc48c656b94e5184519f698cd5386c5a2bf
7
+ File: edgetam.pt
8
+ SHA-256: ed2d4850b8792c239689b043c47046ec239b6e808a3d9b6ae676c803fd8780df
9
+
10
+ Official source and license
11
+ ---------------------------
12
+ Repository: https://github.com/facebookresearch/EdgeTAM
13
+ Commit: 7711e012a30a2402c4eaab637bdb00a521302c91
14
+ License: Apache License 2.0
15
+ LICENSE SHA-256: c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4
16
+
17
+ Conversion implementation
18
+ -------------------------
19
+ Repository: https://github.com/huggingface/transformers
20
+ Commit: bd37c453544e83eb875ed3608980a1660376007a
21
+ File: src/transformers/models/edgetam_video/convert_edgetam_video_to_hf.py
22
+ Copyright 2025 The HuggingFace Inc. team
23
+ License: Apache License 2.0
24
+
25
+ Hash-pinned snapshot metadata source and conversion reference
26
+ -------------------------------------------------------------
27
+ Repository: https://huggingface.co/yonigozlan/EdgeTAM-hf
28
+ Revision: c266ce53b3fc00f0f495b583f6a116c4e57f53bb
29
+ License declared by reference model card: Apache License 2.0
30
+ model.safetensors SHA-256: 8858f8e4757b0b96dab8763f296ecffd845efbbbf698f64163cfa20a63d5fff4
31
+ Files copied byte-for-byte after SHA-256 verification:
32
+ .gitattributes
33
+ config.json
34
+ preprocessor_config.json
35
+ processor_config.json
36
+ video_preprocessor_config.json
37
+
38
+ LibreYOLO modifications
39
+ -----------------------
40
+ The official checkpoint weights were independently safe-loaded and converted
41
+ through lossless key remapping, key/value tensor splitting, and point-embedding
42
+ concatenation; model.safetensors was not copied from the reference. The result
43
+ was strict-loaded into EdgeTamVideoModel. All 984 state tensors were
44
+ checked for exact equality with the pinned reference snapshot. The five
45
+ non-weight files listed above were copied from that reference revision. The raw
46
+ .pt checkpoint was omitted from this mirror.
47
+
48
+ Generated model.safetensors SHA-256: 8858f8e4757b0b96dab8763f296ecffd845efbbbf698f64163cfa20a63d5fff4
README.md ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: transformers
4
+ pipeline_tag: mask-generation
5
+ tags:
6
+ - libreyolo
7
+ - edgetam
8
+ - promptable-segmentation
9
+ - image-segmentation
10
+ base_model: facebook/EdgeTAM
11
+ ---
12
+
13
+ # LibreEdgeTAM
14
+
15
+ This is LibreYOLO's Transformers-compatible mirror of the official EdgeTAM
16
+ checkpoint. The raw pickle-based `.pt` file is deliberately not included.
17
+
18
+ ## Source
19
+
20
+ - Official checkpoint: [`facebook/EdgeTAM`](https://huggingface.co/facebook/EdgeTAM)
21
+ at revision `14d7ecc48c656b94e5184519f698cd5386c5a2bf`
22
+ - `edgetam.pt` SHA-256: `ed2d4850b8792c239689b043c47046ec239b6e808a3d9b6ae676c803fd8780df`
23
+ - Official source: [https://github.com/facebookresearch/EdgeTAM](https://github.com/facebookresearch/EdgeTAM) at commit
24
+ `7711e012a30a2402c4eaab637bdb00a521302c91`
25
+ - Reference Transformers snapshot: [`yonigozlan/EdgeTAM-hf`](https://huggingface.co/yonigozlan/EdgeTAM-hf)
26
+ at revision `c266ce53b3fc00f0f495b583f6a116c4e57f53bb`; its model card declares Apache-2.0
27
+ - Conversion implementation: [`huggingface/transformers`](https://github.com/huggingface/transformers) at
28
+ commit `bd37c453544e83eb875ed3608980a1660376007a`, file `src/transformers/models/edgetam_video/convert_edgetam_video_to_hf.py`
29
+
30
+ ## Modifications
31
+
32
+ LibreYOLO independently converted the model weights from the safely loaded
33
+ official checkpoint using the pinned Apache-2.0 Transformers conversion (key
34
+ remapping, lossless key/value tensor splitting, and point-embedding
35
+ concatenation). It strict-loaded `EdgeTamVideoModel` and checked all
36
+ 984 resulting tensors for exact equality with the pinned reference.
37
+ No learned numeric parameter was changed, and `model.safetensors` was not copied
38
+ from the reference.
39
+
40
+ The non-weight `.gitattributes`, `config.json`, `preprocessor_config.json`,
41
+ `processor_config.json`, and `video_preprocessor_config.json` files are copied
42
+ byte-for-byte from the hash-pinned reference revision above. Every copied file
43
+ is SHA-256 verified. The reference repository declares Apache-2.0 in its model
44
+ card.
45
+
46
+ Generated `model.safetensors` SHA-256: `8858f8e4757b0b96dab8763f296ecffd845efbbbf698f64163cfa20a63d5fff4`.
47
+
48
+ ## Usage
49
+
50
+ ```python
51
+ from transformers import AutoModel
52
+
53
+ model = AutoModel.from_pretrained("LibreYOLO/LibreEdgeTAM", trust_remote_code=False)
54
+ ```
55
+
56
+ LibreYOLO users can load it through `LibreEdgeTAM` once the EdgeTAM integration is
57
+ installed.
58
+
59
+ ## License
60
+
61
+ EdgeTAM code and model checkpoints are licensed under Apache License 2.0. The
62
+ verbatim upstream license is included in `LICENSE`; provenance and modification
63
+ details are included in `NOTICE`.
config.json ADDED
@@ -0,0 +1,1225 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "EdgeTamVideoModel"
4
+ ],
5
+ "dtype": "float32",
6
+ "enable_occlusion_spatial_embedding": false,
7
+ "enable_temporal_pos_encoding_for_object_pointers": false,
8
+ "image_size": 1024,
9
+ "initializer_range": 0.02,
10
+ "mask_decoder_config": {
11
+ "attention_downsample_rate": 2,
12
+ "dynamic_multimask_stability_delta": 0.05,
13
+ "dynamic_multimask_stability_thresh": 0.98,
14
+ "dynamic_multimask_via_stability": true,
15
+ "hidden_act": "gelu",
16
+ "hidden_size": 256,
17
+ "iou_head_depth": 3,
18
+ "iou_head_hidden_dim": 256,
19
+ "mlp_dim": 2048,
20
+ "model_type": "",
21
+ "num_attention_heads": 8,
22
+ "num_hidden_layers": 2,
23
+ "num_multimask_outputs": 3
24
+ },
25
+ "mask_downsampler_embed_dim": 256,
26
+ "mask_downsampler_hidden_act": "gelu",
27
+ "mask_downsampler_kernel_size": 3,
28
+ "mask_downsampler_padding": 1,
29
+ "mask_downsampler_stride": 2,
30
+ "mask_downsampler_total_stride": 16,
31
+ "max_object_pointers_in_encoder": 16,
32
+ "memory_attention_downsample_rate": 1,
33
+ "memory_attention_dropout": 0.1,
34
+ "memory_attention_hidden_size": 256,
35
+ "memory_attention_mlp_hidden_act": "relu",
36
+ "memory_attention_mlp_hidden_size": 2048,
37
+ "memory_attention_num_attention_heads": 1,
38
+ "memory_attention_num_layers": 2,
39
+ "memory_attention_rope_dropout": 0.1,
40
+ "memory_attention_rope_feat_sizes": [
41
+ 64,
42
+ 64
43
+ ],
44
+ "memory_attention_rope_k_sizes": [
45
+ 16,
46
+ 16
47
+ ],
48
+ "memory_attention_rope_theta": 10000,
49
+ "memory_encoder_hidden_size": 256,
50
+ "memory_encoder_output_channels": 64,
51
+ "memory_fuser_embed_dim": 256,
52
+ "memory_fuser_hidden_act": "gelu",
53
+ "memory_fuser_intermediate_dim": 1024,
54
+ "memory_fuser_kernel_size": 7,
55
+ "memory_fuser_layer_scale_init_value": 1e-06,
56
+ "memory_fuser_num_layers": 2,
57
+ "memory_fuser_padding": 3,
58
+ "model_type": "edgetam_video",
59
+ "multimask_max_pt_num": 1,
60
+ "multimask_min_pt_num": 0,
61
+ "multimask_output_for_tracking": true,
62
+ "multimask_output_in_sam": true,
63
+ "num_maskmem": 7,
64
+ "perceiver_resampler_attention_dropout": 0.0,
65
+ "perceiver_resampler_attention_head_dim": 64,
66
+ "perceiver_resampler_hidden_dropout": 0.0,
67
+ "perceiver_resampler_hidden_size": 64,
68
+ "perceiver_resampler_mlp_intermediate_size": 256,
69
+ "perceiver_resampler_num_attention_heads": 1,
70
+ "perceiver_resampler_num_latents": 256,
71
+ "perceiver_resampler_num_latents_2d": 256,
72
+ "perceiver_resampler_num_layers": 2,
73
+ "prompt_encoder_config": {
74
+ "hidden_act": "gelu",
75
+ "hidden_size": 256,
76
+ "image_size": 1024,
77
+ "layer_norm_eps": 1e-06,
78
+ "mask_input_channels": 16,
79
+ "model_type": "",
80
+ "num_point_embeddings": 4,
81
+ "patch_size": 16,
82
+ "scale": 1
83
+ },
84
+ "sigmoid_bias_for_mem_enc": -10.0,
85
+ "sigmoid_scale_for_mem_enc": 20.0,
86
+ "transformers_version": "4.57.0.dev0",
87
+ "vision_config": {
88
+ "backbone_channel_list": [
89
+ 384,
90
+ 192,
91
+ 96,
92
+ 48
93
+ ],
94
+ "backbone_config": {
95
+ "_name_or_path": "",
96
+ "add_cross_attention": false,
97
+ "architecture": "repvit_m1",
98
+ "architectures": null,
99
+ "bad_words_ids": null,
100
+ "begin_suppress_tokens": null,
101
+ "bos_token_id": null,
102
+ "chunk_size_feed_forward": 0,
103
+ "cross_attention_hidden_size": null,
104
+ "decoder_start_token_id": null,
105
+ "diversity_penalty": 0.0,
106
+ "do_pooling": true,
107
+ "do_sample": false,
108
+ "dtype": null,
109
+ "early_stopping": false,
110
+ "encoder_no_repeat_ngram_size": 0,
111
+ "eos_token_id": null,
112
+ "exponential_decay_length_penalty": null,
113
+ "finetuning_task": null,
114
+ "forced_bos_token_id": null,
115
+ "forced_eos_token_id": null,
116
+ "global_pool": "avg",
117
+ "initializer_range": 0.02,
118
+ "is_decoder": false,
119
+ "is_encoder_decoder": false,
120
+ "label_names": [
121
+ "tench, Tinca tinca",
122
+ "goldfish, Carassius auratus",
123
+ "great white shark, white shark, man-eater, man-eating shark, Carcharodon carcharias",
124
+ "tiger shark, Galeocerdo cuvieri",
125
+ "hammerhead, hammerhead shark",
126
+ "electric ray, crampfish, numbfish, torpedo",
127
+ "stingray",
128
+ "cock",
129
+ "hen",
130
+ "ostrich, Struthio camelus",
131
+ "brambling, Fringilla montifringilla",
132
+ "goldfinch, Carduelis carduelis",
133
+ "house finch, linnet, Carpodacus mexicanus",
134
+ "junco, snowbird",
135
+ "indigo bunting, indigo finch, indigo bird, Passerina cyanea",
136
+ "robin, American robin, Turdus migratorius",
137
+ "bulbul",
138
+ "jay",
139
+ "magpie",
140
+ "chickadee",
141
+ "water ouzel, dipper",
142
+ "kite",
143
+ "bald eagle, American eagle, Haliaeetus leucocephalus",
144
+ "vulture",
145
+ "great grey owl, great gray owl, Strix nebulosa",
146
+ "European fire salamander, Salamandra salamandra",
147
+ "common newt, Triturus vulgaris",
148
+ "eft",
149
+ "spotted salamander, Ambystoma maculatum",
150
+ "axolotl, mud puppy, Ambystoma mexicanum",
151
+ "bullfrog, Rana catesbeiana",
152
+ "tree frog, tree-frog",
153
+ "tailed frog, bell toad, ribbed toad, tailed toad, Ascaphus trui",
154
+ "loggerhead, loggerhead turtle, Caretta caretta",
155
+ "leatherback turtle, leatherback, leathery turtle, Dermochelys coriacea",
156
+ "mud turtle",
157
+ "terrapin",
158
+ "box turtle, box tortoise",
159
+ "banded gecko",
160
+ "common iguana, iguana, Iguana iguana",
161
+ "American chameleon, anole, Anolis carolinensis",
162
+ "whiptail, whiptail lizard",
163
+ "agama",
164
+ "frilled lizard, Chlamydosaurus kingi",
165
+ "alligator lizard",
166
+ "Gila monster, Heloderma suspectum",
167
+ "green lizard, Lacerta viridis",
168
+ "African chameleon, Chamaeleo chamaeleon",
169
+ "Komodo dragon, Komodo lizard, dragon lizard, giant lizard, Varanus komodoensis",
170
+ "African crocodile, Nile crocodile, Crocodylus niloticus",
171
+ "American alligator, Alligator mississipiensis",
172
+ "triceratops",
173
+ "thunder snake, worm snake, Carphophis amoenus",
174
+ "ringneck snake, ring-necked snake, ring snake",
175
+ "hognose snake, puff adder, sand viper",
176
+ "green snake, grass snake",
177
+ "king snake, kingsnake",
178
+ "garter snake, grass snake",
179
+ "water snake",
180
+ "vine snake",
181
+ "night snake, Hypsiglena torquata",
182
+ "boa constrictor, Constrictor constrictor",
183
+ "rock python, rock snake, Python sebae",
184
+ "Indian cobra, Naja naja",
185
+ "green mamba",
186
+ "sea snake",
187
+ "horned viper, cerastes, sand viper, horned asp, Cerastes cornutus",
188
+ "diamondback, diamondback rattlesnake, Crotalus adamanteus",
189
+ "sidewinder, horned rattlesnake, Crotalus cerastes",
190
+ "trilobite",
191
+ "harvestman, daddy longlegs, Phalangium opilio",
192
+ "scorpion",
193
+ "black and gold garden spider, Argiope aurantia",
194
+ "barn spider, Araneus cavaticus",
195
+ "garden spider, Aranea diademata",
196
+ "black widow, Latrodectus mactans",
197
+ "tarantula",
198
+ "wolf spider, hunting spider",
199
+ "tick",
200
+ "centipede",
201
+ "black grouse",
202
+ "ptarmigan",
203
+ "ruffed grouse, partridge, Bonasa umbellus",
204
+ "prairie chicken, prairie grouse, prairie fowl",
205
+ "peacock",
206
+ "quail",
207
+ "partridge",
208
+ "African grey, African gray, Psittacus erithacus",
209
+ "macaw",
210
+ "sulphur-crested cockatoo, Kakatoe galerita, Cacatua galerita",
211
+ "lorikeet",
212
+ "coucal",
213
+ "bee eater",
214
+ "hornbill",
215
+ "hummingbird",
216
+ "jacamar",
217
+ "toucan",
218
+ "drake",
219
+ "red-breasted merganser, Mergus serrator",
220
+ "goose",
221
+ "black swan, Cygnus atratus",
222
+ "tusker",
223
+ "echidna, spiny anteater, anteater",
224
+ "platypus, duckbill, duckbilled platypus, duck-billed platypus, Ornithorhynchus anatinus",
225
+ "wallaby, brush kangaroo",
226
+ "koala, koala bear, kangaroo bear, native bear, Phascolarctos cinereus",
227
+ "wombat",
228
+ "jellyfish",
229
+ "sea anemone, anemone",
230
+ "brain coral",
231
+ "flatworm, platyhelminth",
232
+ "nematode, nematode worm, roundworm",
233
+ "conch",
234
+ "snail",
235
+ "slug",
236
+ "sea slug, nudibranch",
237
+ "chiton, coat-of-mail shell, sea cradle, polyplacophore",
238
+ "chambered nautilus, pearly nautilus, nautilus",
239
+ "Dungeness crab, Cancer magister",
240
+ "rock crab, Cancer irroratus",
241
+ "fiddler crab",
242
+ "king crab, Alaska crab, Alaskan king crab, Alaska king crab, Paralithodes camtschatica",
243
+ "American lobster, Northern lobster, Maine lobster, Homarus americanus",
244
+ "spiny lobster, langouste, rock lobster, crawfish, crayfish, sea crawfish",
245
+ "crayfish, crawfish, crawdad, crawdaddy",
246
+ "hermit crab",
247
+ "isopod",
248
+ "white stork, Ciconia ciconia",
249
+ "black stork, Ciconia nigra",
250
+ "spoonbill",
251
+ "flamingo",
252
+ "little blue heron, Egretta caerulea",
253
+ "American egret, great white heron, Egretta albus",
254
+ "bittern",
255
+ "crane",
256
+ "limpkin, Aramus pictus",
257
+ "European gallinule, Porphyrio porphyrio",
258
+ "American coot, marsh hen, mud hen, water hen, Fulica americana",
259
+ "bustard",
260
+ "ruddy turnstone, Arenaria interpres",
261
+ "red-backed sandpiper, dunlin, Erolia alpina",
262
+ "redshank, Tringa totanus",
263
+ "dowitcher",
264
+ "oystercatcher, oyster catcher",
265
+ "pelican",
266
+ "king penguin, Aptenodytes patagonica",
267
+ "albatross, mollymawk",
268
+ "grey whale, gray whale, devilfish, Eschrichtius gibbosus, Eschrichtius robustus",
269
+ "killer whale, killer, orca, grampus, sea wolf, Orcinus orca",
270
+ "dugong, Dugong dugon",
271
+ "sea lion",
272
+ "Chihuahua",
273
+ "Japanese spaniel",
274
+ "Maltese dog, Maltese terrier, Maltese",
275
+ "Pekinese, Pekingese, Peke",
276
+ "Shih-Tzu",
277
+ "Blenheim spaniel",
278
+ "papillon",
279
+ "toy terrier",
280
+ "Rhodesian ridgeback",
281
+ "Afghan hound, Afghan",
282
+ "basset, basset hound",
283
+ "beagle",
284
+ "bloodhound, sleuthhound",
285
+ "bluetick",
286
+ "black-and-tan coonhound",
287
+ "Walker hound, Walker foxhound",
288
+ "English foxhound",
289
+ "redbone",
290
+ "borzoi, Russian wolfhound",
291
+ "Irish wolfhound",
292
+ "Italian greyhound",
293
+ "whippet",
294
+ "Ibizan hound, Ibizan Podenco",
295
+ "Norwegian elkhound, elkhound",
296
+ "otterhound, otter hound",
297
+ "Saluki, gazelle hound",
298
+ "Scottish deerhound, deerhound",
299
+ "Weimaraner",
300
+ "Staffordshire bullterrier, Staffordshire bull terrier",
301
+ "American Staffordshire terrier, Staffordshire terrier, American pit bull terrier, pit bull terrier",
302
+ "Bedlington terrier",
303
+ "Border terrier",
304
+ "Kerry blue terrier",
305
+ "Irish terrier",
306
+ "Norfolk terrier",
307
+ "Norwich terrier",
308
+ "Yorkshire terrier",
309
+ "wire-haired fox terrier",
310
+ "Lakeland terrier",
311
+ "Sealyham terrier, Sealyham",
312
+ "Airedale, Airedale terrier",
313
+ "cairn, cairn terrier",
314
+ "Australian terrier",
315
+ "Dandie Dinmont, Dandie Dinmont terrier",
316
+ "Boston bull, Boston terrier",
317
+ "miniature schnauzer",
318
+ "giant schnauzer",
319
+ "standard schnauzer",
320
+ "Scotch terrier, Scottish terrier, Scottie",
321
+ "Tibetan terrier, chrysanthemum dog",
322
+ "silky terrier, Sydney silky",
323
+ "soft-coated wheaten terrier",
324
+ "West Highland white terrier",
325
+ "Lhasa, Lhasa apso",
326
+ "flat-coated retriever",
327
+ "curly-coated retriever",
328
+ "golden retriever",
329
+ "Labrador retriever",
330
+ "Chesapeake Bay retriever",
331
+ "German short-haired pointer",
332
+ "vizsla, Hungarian pointer",
333
+ "English setter",
334
+ "Irish setter, red setter",
335
+ "Gordon setter",
336
+ "Brittany spaniel",
337
+ "clumber, clumber spaniel",
338
+ "English springer, English springer spaniel",
339
+ "Welsh springer spaniel",
340
+ "cocker spaniel, English cocker spaniel, cocker",
341
+ "Sussex spaniel",
342
+ "Irish water spaniel",
343
+ "kuvasz",
344
+ "schipperke",
345
+ "groenendael",
346
+ "malinois",
347
+ "briard",
348
+ "kelpie",
349
+ "komondor",
350
+ "Old English sheepdog, bobtail",
351
+ "Shetland sheepdog, Shetland sheep dog, Shetland",
352
+ "collie",
353
+ "Border collie",
354
+ "Bouvier des Flandres, Bouviers des Flandres",
355
+ "Rottweiler",
356
+ "German shepherd, German shepherd dog, German police dog, alsatian",
357
+ "Doberman, Doberman pinscher",
358
+ "miniature pinscher",
359
+ "Greater Swiss Mountain dog",
360
+ "Bernese mountain dog",
361
+ "Appenzeller",
362
+ "EntleBucher",
363
+ "boxer",
364
+ "bull mastiff",
365
+ "Tibetan mastiff",
366
+ "French bulldog",
367
+ "Great Dane",
368
+ "Saint Bernard, St Bernard",
369
+ "Eskimo dog, husky",
370
+ "malamute, malemute, Alaskan malamute",
371
+ "Siberian husky",
372
+ "dalmatian, coach dog, carriage dog",
373
+ "affenpinscher, monkey pinscher, monkey dog",
374
+ "basenji",
375
+ "pug, pug-dog",
376
+ "Leonberg",
377
+ "Newfoundland, Newfoundland dog",
378
+ "Great Pyrenees",
379
+ "Samoyed, Samoyede",
380
+ "Pomeranian",
381
+ "chow, chow chow",
382
+ "keeshond",
383
+ "Brabancon griffon",
384
+ "Pembroke, Pembroke Welsh corgi",
385
+ "Cardigan, Cardigan Welsh corgi",
386
+ "toy poodle",
387
+ "miniature poodle",
388
+ "standard poodle",
389
+ "Mexican hairless",
390
+ "timber wolf, grey wolf, gray wolf, Canis lupus",
391
+ "white wolf, Arctic wolf, Canis lupus tundrarum",
392
+ "red wolf, maned wolf, Canis rufus, Canis niger",
393
+ "coyote, prairie wolf, brush wolf, Canis latrans",
394
+ "dingo, warrigal, warragal, Canis dingo",
395
+ "dhole, Cuon alpinus",
396
+ "African hunting dog, hyena dog, Cape hunting dog, Lycaon pictus",
397
+ "hyena, hyaena",
398
+ "red fox, Vulpes vulpes",
399
+ "kit fox, Vulpes macrotis",
400
+ "Arctic fox, white fox, Alopex lagopus",
401
+ "grey fox, gray fox, Urocyon cinereoargenteus",
402
+ "tabby, tabby cat",
403
+ "tiger cat",
404
+ "Persian cat",
405
+ "Siamese cat, Siamese",
406
+ "Egyptian cat",
407
+ "cougar, puma, catamount, mountain lion, painter, panther, Felis concolor",
408
+ "lynx, catamount",
409
+ "leopard, Panthera pardus",
410
+ "snow leopard, ounce, Panthera uncia",
411
+ "jaguar, panther, Panthera onca, Felis onca",
412
+ "lion, king of beasts, Panthera leo",
413
+ "tiger, Panthera tigris",
414
+ "cheetah, chetah, Acinonyx jubatus",
415
+ "brown bear, bruin, Ursus arctos",
416
+ "American black bear, black bear, Ursus americanus, Euarctos americanus",
417
+ "ice bear, polar bear, Ursus Maritimus, Thalarctos maritimus",
418
+ "sloth bear, Melursus ursinus, Ursus ursinus",
419
+ "mongoose",
420
+ "meerkat, mierkat",
421
+ "tiger beetle",
422
+ "ladybug, ladybeetle, lady beetle, ladybird, ladybird beetle",
423
+ "ground beetle, carabid beetle",
424
+ "long-horned beetle, longicorn, longicorn beetle",
425
+ "leaf beetle, chrysomelid",
426
+ "dung beetle",
427
+ "rhinoceros beetle",
428
+ "weevil",
429
+ "fly",
430
+ "bee",
431
+ "ant, emmet, pismire",
432
+ "grasshopper, hopper",
433
+ "cricket",
434
+ "walking stick, walkingstick, stick insect",
435
+ "cockroach, roach",
436
+ "mantis, mantid",
437
+ "cicada, cicala",
438
+ "leafhopper",
439
+ "lacewing, lacewing fly",
440
+ "dragonfly, darning needle, devil's darning needle, sewing needle, snake feeder, snake doctor, mosquito hawk, skeeter hawk",
441
+ "damselfly",
442
+ "admiral",
443
+ "ringlet, ringlet butterfly",
444
+ "monarch, monarch butterfly, milkweed butterfly, Danaus plexippus",
445
+ "cabbage butterfly",
446
+ "sulphur butterfly, sulfur butterfly",
447
+ "lycaenid, lycaenid butterfly",
448
+ "starfish, sea star",
449
+ "sea urchin",
450
+ "sea cucumber, holothurian",
451
+ "wood rabbit, cottontail, cottontail rabbit",
452
+ "hare",
453
+ "Angora, Angora rabbit",
454
+ "hamster",
455
+ "porcupine, hedgehog",
456
+ "fox squirrel, eastern fox squirrel, Sciurus niger",
457
+ "marmot",
458
+ "beaver",
459
+ "guinea pig, Cavia cobaya",
460
+ "sorrel",
461
+ "zebra",
462
+ "hog, pig, grunter, squealer, Sus scrofa",
463
+ "wild boar, boar, Sus scrofa",
464
+ "warthog",
465
+ "hippopotamus, hippo, river horse, Hippopotamus amphibius",
466
+ "ox",
467
+ "water buffalo, water ox, Asiatic buffalo, Bubalus bubalis",
468
+ "bison",
469
+ "ram, tup",
470
+ "bighorn, bighorn sheep, cimarron, Rocky Mountain bighorn, Rocky Mountain sheep, Ovis canadensis",
471
+ "ibex, Capra ibex",
472
+ "hartebeest",
473
+ "impala, Aepyceros melampus",
474
+ "gazelle",
475
+ "Arabian camel, dromedary, Camelus dromedarius",
476
+ "llama",
477
+ "weasel",
478
+ "mink",
479
+ "polecat, fitch, foulmart, foumart, Mustela putorius",
480
+ "black-footed ferret, ferret, Mustela nigripes",
481
+ "otter",
482
+ "skunk, polecat, wood pussy",
483
+ "badger",
484
+ "armadillo",
485
+ "three-toed sloth, ai, Bradypus tridactylus",
486
+ "orangutan, orang, orangutang, Pongo pygmaeus",
487
+ "gorilla, Gorilla gorilla",
488
+ "chimpanzee, chimp, Pan troglodytes",
489
+ "gibbon, Hylobates lar",
490
+ "siamang, Hylobates syndactylus, Symphalangus syndactylus",
491
+ "guenon, guenon monkey",
492
+ "patas, hussar monkey, Erythrocebus patas",
493
+ "baboon",
494
+ "macaque",
495
+ "langur",
496
+ "colobus, colobus monkey",
497
+ "proboscis monkey, Nasalis larvatus",
498
+ "marmoset",
499
+ "capuchin, ringtail, Cebus capucinus",
500
+ "howler monkey, howler",
501
+ "titi, titi monkey",
502
+ "spider monkey, Ateles geoffroyi",
503
+ "squirrel monkey, Saimiri sciureus",
504
+ "Madagascar cat, ring-tailed lemur, Lemur catta",
505
+ "indri, indris, Indri indri, Indri brevicaudatus",
506
+ "Indian elephant, Elephas maximus",
507
+ "African elephant, Loxodonta africana",
508
+ "lesser panda, red panda, panda, bear cat, cat bear, Ailurus fulgens",
509
+ "giant panda, panda, panda bear, coon bear, Ailuropoda melanoleuca",
510
+ "barracouta, snoek",
511
+ "eel",
512
+ "coho, cohoe, coho salmon, blue jack, silver salmon, Oncorhynchus kisutch",
513
+ "rock beauty, Holocanthus tricolor",
514
+ "anemone fish",
515
+ "sturgeon",
516
+ "gar, garfish, garpike, billfish, Lepisosteus osseus",
517
+ "lionfish",
518
+ "puffer, pufferfish, blowfish, globefish",
519
+ "abacus",
520
+ "abaya",
521
+ "academic gown, academic robe, judge's robe",
522
+ "accordion, piano accordion, squeeze box",
523
+ "acoustic guitar",
524
+ "aircraft carrier, carrier, flattop, attack aircraft carrier",
525
+ "airliner",
526
+ "airship, dirigible",
527
+ "altar",
528
+ "ambulance",
529
+ "amphibian, amphibious vehicle",
530
+ "analog clock",
531
+ "apiary, bee house",
532
+ "apron",
533
+ "ashcan, trash can, garbage can, wastebin, ash bin, ash-bin, ashbin, dustbin, trash barrel, trash bin",
534
+ "assault rifle, assault gun",
535
+ "backpack, back pack, knapsack, packsack, rucksack, haversack",
536
+ "bakery, bakeshop, bakehouse",
537
+ "balance beam, beam",
538
+ "balloon",
539
+ "ballpoint, ballpoint pen, ballpen, Biro",
540
+ "Band Aid",
541
+ "banjo",
542
+ "bannister, banister, balustrade, balusters, handrail",
543
+ "barbell",
544
+ "barber chair",
545
+ "barbershop",
546
+ "barn",
547
+ "barometer",
548
+ "barrel, cask",
549
+ "barrow, garden cart, lawn cart, wheelbarrow",
550
+ "baseball",
551
+ "basketball",
552
+ "bassinet",
553
+ "bassoon",
554
+ "bathing cap, swimming cap",
555
+ "bath towel",
556
+ "bathtub, bathing tub, bath, tub",
557
+ "beach wagon, station wagon, wagon, estate car, beach waggon, station waggon, waggon",
558
+ "beacon, lighthouse, beacon light, pharos",
559
+ "beaker",
560
+ "bearskin, busby, shako",
561
+ "beer bottle",
562
+ "beer glass",
563
+ "bell cote, bell cot",
564
+ "bib",
565
+ "bicycle-built-for-two, tandem bicycle, tandem",
566
+ "bikini, two-piece",
567
+ "binder, ring-binder",
568
+ "binoculars, field glasses, opera glasses",
569
+ "birdhouse",
570
+ "boathouse",
571
+ "bobsled, bobsleigh, bob",
572
+ "bolo tie, bolo, bola tie, bola",
573
+ "bonnet, poke bonnet",
574
+ "bookcase",
575
+ "bookshop, bookstore, bookstall",
576
+ "bottlecap",
577
+ "bow",
578
+ "bow tie, bow-tie, bowtie",
579
+ "brass, memorial tablet, plaque",
580
+ "brassiere, bra, bandeau",
581
+ "breakwater, groin, groyne, mole, bulwark, seawall, jetty",
582
+ "breastplate, aegis, egis",
583
+ "broom",
584
+ "bucket, pail",
585
+ "buckle",
586
+ "bulletproof vest",
587
+ "bullet train, bullet",
588
+ "butcher shop, meat market",
589
+ "cab, hack, taxi, taxicab",
590
+ "caldron, cauldron",
591
+ "candle, taper, wax light",
592
+ "cannon",
593
+ "canoe",
594
+ "can opener, tin opener",
595
+ "cardigan",
596
+ "car mirror",
597
+ "carousel, carrousel, merry-go-round, roundabout, whirligig",
598
+ "carpenter's kit, tool kit",
599
+ "carton",
600
+ "car wheel",
601
+ "cash machine, cash dispenser, automated teller machine, automatic teller machine, automated teller, automatic teller, ATM",
602
+ "cassette",
603
+ "cassette player",
604
+ "castle",
605
+ "catamaran",
606
+ "CD player",
607
+ "cello, violoncello",
608
+ "cellular telephone, cellular phone, cellphone, cell, mobile phone",
609
+ "chain",
610
+ "chainlink fence",
611
+ "chain mail, ring mail, mail, chain armor, chain armour, ring armor, ring armour",
612
+ "chain saw, chainsaw",
613
+ "chest",
614
+ "chiffonier, commode",
615
+ "chime, bell, gong",
616
+ "china cabinet, china closet",
617
+ "Christmas stocking",
618
+ "church, church building",
619
+ "cinema, movie theater, movie theatre, movie house, picture palace",
620
+ "cleaver, meat cleaver, chopper",
621
+ "cliff dwelling",
622
+ "cloak",
623
+ "clog, geta, patten, sabot",
624
+ "cocktail shaker",
625
+ "coffee mug",
626
+ "coffeepot",
627
+ "coil, spiral, volute, whorl, helix",
628
+ "combination lock",
629
+ "computer keyboard, keypad",
630
+ "confectionery, confectionary, candy store",
631
+ "container ship, containership, container vessel",
632
+ "convertible",
633
+ "corkscrew, bottle screw",
634
+ "cornet, horn, trumpet, trump",
635
+ "cowboy boot",
636
+ "cowboy hat, ten-gallon hat",
637
+ "cradle",
638
+ "crane",
639
+ "crash helmet",
640
+ "crate",
641
+ "crib, cot",
642
+ "Crock Pot",
643
+ "croquet ball",
644
+ "crutch",
645
+ "cuirass",
646
+ "dam, dike, dyke",
647
+ "desk",
648
+ "desktop computer",
649
+ "dial telephone, dial phone",
650
+ "diaper, nappy, napkin",
651
+ "digital clock",
652
+ "digital watch",
653
+ "dining table, board",
654
+ "dishrag, dishcloth",
655
+ "dishwasher, dish washer, dishwashing machine",
656
+ "disk brake, disc brake",
657
+ "dock, dockage, docking facility",
658
+ "dogsled, dog sled, dog sleigh",
659
+ "dome",
660
+ "doormat, welcome mat",
661
+ "drilling platform, offshore rig",
662
+ "drum, membranophone, tympan",
663
+ "drumstick",
664
+ "dumbbell",
665
+ "Dutch oven",
666
+ "electric fan, blower",
667
+ "electric guitar",
668
+ "electric locomotive",
669
+ "entertainment center",
670
+ "envelope",
671
+ "espresso maker",
672
+ "face powder",
673
+ "feather boa, boa",
674
+ "file, file cabinet, filing cabinet",
675
+ "fireboat",
676
+ "fire engine, fire truck",
677
+ "fire screen, fireguard",
678
+ "flagpole, flagstaff",
679
+ "flute, transverse flute",
680
+ "folding chair",
681
+ "football helmet",
682
+ "forklift",
683
+ "fountain",
684
+ "fountain pen",
685
+ "four-poster",
686
+ "freight car",
687
+ "French horn, horn",
688
+ "frying pan, frypan, skillet",
689
+ "fur coat",
690
+ "garbage truck, dustcart",
691
+ "gasmask, respirator, gas helmet",
692
+ "gas pump, gasoline pump, petrol pump, island dispenser",
693
+ "goblet",
694
+ "go-kart",
695
+ "golf ball",
696
+ "golfcart, golf cart",
697
+ "gondola",
698
+ "gong, tam-tam",
699
+ "gown",
700
+ "grand piano, grand",
701
+ "greenhouse, nursery, glasshouse",
702
+ "grille, radiator grille",
703
+ "grocery store, grocery, food market, market",
704
+ "guillotine",
705
+ "hair slide",
706
+ "hair spray",
707
+ "half track",
708
+ "hammer",
709
+ "hamper",
710
+ "hand blower, blow dryer, blow drier, hair dryer, hair drier",
711
+ "hand-held computer, hand-held microcomputer",
712
+ "handkerchief, hankie, hanky, hankey",
713
+ "hard disc, hard disk, fixed disk",
714
+ "harmonica, mouth organ, harp, mouth harp",
715
+ "harp",
716
+ "harvester, reaper",
717
+ "hatchet",
718
+ "holster",
719
+ "home theater, home theatre",
720
+ "honeycomb",
721
+ "hook, claw",
722
+ "hoopskirt, crinoline",
723
+ "horizontal bar, high bar",
724
+ "horse cart, horse-cart",
725
+ "hourglass",
726
+ "iPod",
727
+ "iron, smoothing iron",
728
+ "jack-o'-lantern",
729
+ "jean, blue jean, denim",
730
+ "jeep, landrover",
731
+ "jersey, T-shirt, tee shirt",
732
+ "jigsaw puzzle",
733
+ "jinrikisha, ricksha, rickshaw",
734
+ "joystick",
735
+ "kimono",
736
+ "knee pad",
737
+ "knot",
738
+ "lab coat, laboratory coat",
739
+ "ladle",
740
+ "lampshade, lamp shade",
741
+ "laptop, laptop computer",
742
+ "lawn mower, mower",
743
+ "lens cap, lens cover",
744
+ "letter opener, paper knife, paperknife",
745
+ "library",
746
+ "lifeboat",
747
+ "lighter, light, igniter, ignitor",
748
+ "limousine, limo",
749
+ "liner, ocean liner",
750
+ "lipstick, lip rouge",
751
+ "Loafer",
752
+ "lotion",
753
+ "loudspeaker, speaker, speaker unit, loudspeaker system, speaker system",
754
+ "loupe, jeweler's loupe",
755
+ "lumbermill, sawmill",
756
+ "magnetic compass",
757
+ "mailbag, postbag",
758
+ "mailbox, letter box",
759
+ "maillot",
760
+ "maillot, tank suit",
761
+ "manhole cover",
762
+ "maraca",
763
+ "marimba, xylophone",
764
+ "mask",
765
+ "matchstick",
766
+ "maypole",
767
+ "maze, labyrinth",
768
+ "measuring cup",
769
+ "medicine chest, medicine cabinet",
770
+ "megalith, megalithic structure",
771
+ "microphone, mike",
772
+ "microwave, microwave oven",
773
+ "military uniform",
774
+ "milk can",
775
+ "minibus",
776
+ "miniskirt, mini",
777
+ "minivan",
778
+ "missile",
779
+ "mitten",
780
+ "mixing bowl",
781
+ "mobile home, manufactured home",
782
+ "Model T",
783
+ "modem",
784
+ "monastery",
785
+ "monitor",
786
+ "moped",
787
+ "mortar",
788
+ "mortarboard",
789
+ "mosque",
790
+ "mosquito net",
791
+ "motor scooter, scooter",
792
+ "mountain bike, all-terrain bike, off-roader",
793
+ "mountain tent",
794
+ "mouse, computer mouse",
795
+ "mousetrap",
796
+ "moving van",
797
+ "muzzle",
798
+ "nail",
799
+ "neck brace",
800
+ "necklace",
801
+ "nipple",
802
+ "notebook, notebook computer",
803
+ "obelisk",
804
+ "oboe, hautboy, hautbois",
805
+ "ocarina, sweet potato",
806
+ "odometer, hodometer, mileometer, milometer",
807
+ "oil filter",
808
+ "organ, pipe organ",
809
+ "oscilloscope, scope, cathode-ray oscilloscope, CRO",
810
+ "overskirt",
811
+ "oxcart",
812
+ "oxygen mask",
813
+ "packet",
814
+ "paddle, boat paddle",
815
+ "paddlewheel, paddle wheel",
816
+ "padlock",
817
+ "paintbrush",
818
+ "pajama, pyjama, pj's, jammies",
819
+ "palace",
820
+ "panpipe, pandean pipe, syrinx",
821
+ "paper towel",
822
+ "parachute, chute",
823
+ "parallel bars, bars",
824
+ "park bench",
825
+ "parking meter",
826
+ "passenger car, coach, carriage",
827
+ "patio, terrace",
828
+ "pay-phone, pay-station",
829
+ "pedestal, plinth, footstall",
830
+ "pencil box, pencil case",
831
+ "pencil sharpener",
832
+ "perfume, essence",
833
+ "Petri dish",
834
+ "photocopier",
835
+ "pick, plectrum, plectron",
836
+ "pickelhaube",
837
+ "picket fence, paling",
838
+ "pickup, pickup truck",
839
+ "pier",
840
+ "piggy bank, penny bank",
841
+ "pill bottle",
842
+ "pillow",
843
+ "ping-pong ball",
844
+ "pinwheel",
845
+ "pirate, pirate ship",
846
+ "pitcher, ewer",
847
+ "plane, carpenter's plane, woodworking plane",
848
+ "planetarium",
849
+ "plastic bag",
850
+ "plate rack",
851
+ "plow, plough",
852
+ "plunger, plumber's helper",
853
+ "Polaroid camera, Polaroid Land camera",
854
+ "pole",
855
+ "police van, police wagon, paddy wagon, patrol wagon, wagon, black Maria",
856
+ "poncho",
857
+ "pool table, billiard table, snooker table",
858
+ "pop bottle, soda bottle",
859
+ "pot, flowerpot",
860
+ "potter's wheel",
861
+ "power drill",
862
+ "prayer rug, prayer mat",
863
+ "printer",
864
+ "prison, prison house",
865
+ "projectile, missile",
866
+ "projector",
867
+ "puck, hockey puck",
868
+ "punching bag, punch bag, punching ball, punchball",
869
+ "purse",
870
+ "quill, quill pen",
871
+ "quilt, comforter, comfort, puff",
872
+ "racer, race car, racing car",
873
+ "racket, racquet",
874
+ "radiator",
875
+ "radio, wireless",
876
+ "radio telescope, radio reflector",
877
+ "rain barrel",
878
+ "recreational vehicle, RV, R.V.",
879
+ "reel",
880
+ "reflex camera",
881
+ "refrigerator, icebox",
882
+ "remote control, remote",
883
+ "restaurant, eating house, eating place, eatery",
884
+ "revolver, six-gun, six-shooter",
885
+ "rifle",
886
+ "rocking chair, rocker",
887
+ "rotisserie",
888
+ "rubber eraser, rubber, pencil eraser",
889
+ "rugby ball",
890
+ "rule, ruler",
891
+ "running shoe",
892
+ "safe",
893
+ "safety pin",
894
+ "saltshaker, salt shaker",
895
+ "sandal",
896
+ "sarong",
897
+ "sax, saxophone",
898
+ "scabbard",
899
+ "scale, weighing machine",
900
+ "school bus",
901
+ "schooner",
902
+ "scoreboard",
903
+ "screen, CRT screen",
904
+ "screw",
905
+ "screwdriver",
906
+ "seat belt, seatbelt",
907
+ "sewing machine",
908
+ "shield, buckler",
909
+ "shoe shop, shoe-shop, shoe store",
910
+ "shoji",
911
+ "shopping basket",
912
+ "shopping cart",
913
+ "shovel",
914
+ "shower cap",
915
+ "shower curtain",
916
+ "ski",
917
+ "ski mask",
918
+ "sleeping bag",
919
+ "slide rule, slipstick",
920
+ "sliding door",
921
+ "slot, one-armed bandit",
922
+ "snorkel",
923
+ "snowmobile",
924
+ "snowplow, snowplough",
925
+ "soap dispenser",
926
+ "soccer ball",
927
+ "sock",
928
+ "solar dish, solar collector, solar furnace",
929
+ "sombrero",
930
+ "soup bowl",
931
+ "space bar",
932
+ "space heater",
933
+ "space shuttle",
934
+ "spatula",
935
+ "speedboat",
936
+ "spider web, spider's web",
937
+ "spindle",
938
+ "sports car, sport car",
939
+ "spotlight, spot",
940
+ "stage",
941
+ "steam locomotive",
942
+ "steel arch bridge",
943
+ "steel drum",
944
+ "stethoscope",
945
+ "stole",
946
+ "stone wall",
947
+ "stopwatch, stop watch",
948
+ "stove",
949
+ "strainer",
950
+ "streetcar, tram, tramcar, trolley, trolley car",
951
+ "stretcher",
952
+ "studio couch, day bed",
953
+ "stupa, tope",
954
+ "submarine, pigboat, sub, U-boat",
955
+ "suit, suit of clothes",
956
+ "sundial",
957
+ "sunglass",
958
+ "sunglasses, dark glasses, shades",
959
+ "sunscreen, sunblock, sun blocker",
960
+ "suspension bridge",
961
+ "swab, swob, mop",
962
+ "sweatshirt",
963
+ "swimming trunks, bathing trunks",
964
+ "swing",
965
+ "switch, electric switch, electrical switch",
966
+ "syringe",
967
+ "table lamp",
968
+ "tank, army tank, armored combat vehicle, armoured combat vehicle",
969
+ "tape player",
970
+ "teapot",
971
+ "teddy, teddy bear",
972
+ "television, television system",
973
+ "tennis ball",
974
+ "thatch, thatched roof",
975
+ "theater curtain, theatre curtain",
976
+ "thimble",
977
+ "thresher, thrasher, threshing machine",
978
+ "throne",
979
+ "tile roof",
980
+ "toaster",
981
+ "tobacco shop, tobacconist shop, tobacconist",
982
+ "toilet seat",
983
+ "torch",
984
+ "totem pole",
985
+ "tow truck, tow car, wrecker",
986
+ "toyshop",
987
+ "tractor",
988
+ "trailer truck, tractor trailer, trucking rig, rig, articulated lorry, semi",
989
+ "tray",
990
+ "trench coat",
991
+ "tricycle, trike, velocipede",
992
+ "trimaran",
993
+ "tripod",
994
+ "triumphal arch",
995
+ "trolleybus, trolley coach, trackless trolley",
996
+ "trombone",
997
+ "tub, vat",
998
+ "turnstile",
999
+ "typewriter keyboard",
1000
+ "umbrella",
1001
+ "unicycle, monocycle",
1002
+ "upright, upright piano",
1003
+ "vacuum, vacuum cleaner",
1004
+ "vase",
1005
+ "vault",
1006
+ "velvet",
1007
+ "vending machine",
1008
+ "vestment",
1009
+ "viaduct",
1010
+ "violin, fiddle",
1011
+ "volleyball",
1012
+ "waffle iron",
1013
+ "wall clock",
1014
+ "wallet, billfold, notecase, pocketbook",
1015
+ "wardrobe, closet, press",
1016
+ "warplane, military plane",
1017
+ "washbasin, handbasin, washbowl, lavabo, wash-hand basin",
1018
+ "washer, automatic washer, washing machine",
1019
+ "water bottle",
1020
+ "water jug",
1021
+ "water tower",
1022
+ "whiskey jug",
1023
+ "whistle",
1024
+ "wig",
1025
+ "window screen",
1026
+ "window shade",
1027
+ "Windsor tie",
1028
+ "wine bottle",
1029
+ "wing",
1030
+ "wok",
1031
+ "wooden spoon",
1032
+ "wool, woolen, woollen",
1033
+ "worm fence, snake fence, snake-rail fence, Virginia fence",
1034
+ "wreck",
1035
+ "yawl",
1036
+ "yurt",
1037
+ "web site, website, internet site, site",
1038
+ "comic book",
1039
+ "crossword puzzle, crossword",
1040
+ "street sign",
1041
+ "traffic light, traffic signal, stoplight",
1042
+ "book jacket, dust cover, dust jacket, dust wrapper",
1043
+ "menu",
1044
+ "plate",
1045
+ "guacamole",
1046
+ "consomme",
1047
+ "hot pot, hotpot",
1048
+ "trifle",
1049
+ "ice cream, icecream",
1050
+ "ice lolly, lolly, lollipop, popsicle",
1051
+ "French loaf",
1052
+ "bagel, beigel",
1053
+ "pretzel",
1054
+ "cheeseburger",
1055
+ "hotdog, hot dog, red hot",
1056
+ "mashed potato",
1057
+ "head cabbage",
1058
+ "broccoli",
1059
+ "cauliflower",
1060
+ "zucchini, courgette",
1061
+ "spaghetti squash",
1062
+ "acorn squash",
1063
+ "butternut squash",
1064
+ "cucumber, cuke",
1065
+ "artichoke, globe artichoke",
1066
+ "bell pepper",
1067
+ "cardoon",
1068
+ "mushroom",
1069
+ "Granny Smith",
1070
+ "strawberry",
1071
+ "orange",
1072
+ "lemon",
1073
+ "fig",
1074
+ "pineapple, ananas",
1075
+ "banana",
1076
+ "jackfruit, jak, jack",
1077
+ "custard apple",
1078
+ "pomegranate",
1079
+ "hay",
1080
+ "carbonara",
1081
+ "chocolate sauce, chocolate syrup",
1082
+ "dough",
1083
+ "meat loaf, meatloaf",
1084
+ "pizza, pizza pie",
1085
+ "potpie",
1086
+ "burrito",
1087
+ "red wine",
1088
+ "espresso",
1089
+ "cup",
1090
+ "eggnog",
1091
+ "alp",
1092
+ "bubble",
1093
+ "cliff, drop, drop-off",
1094
+ "coral reef",
1095
+ "geyser",
1096
+ "lakeside, lakeshore",
1097
+ "promontory, headland, head, foreland",
1098
+ "sandbar, sand bar",
1099
+ "seashore, coast, seacoast, sea-coast",
1100
+ "valley, vale",
1101
+ "volcano",
1102
+ "ballplayer, baseball player",
1103
+ "groom, bridegroom",
1104
+ "scuba diver",
1105
+ "rapeseed",
1106
+ "daisy",
1107
+ "yellow lady's slipper, yellow lady-slipper, Cypripedium calceolus, Cypripedium parviflorum",
1108
+ "corn",
1109
+ "acorn",
1110
+ "hip, rose hip, rosehip",
1111
+ "buckeye, horse chestnut, conker",
1112
+ "coral fungus",
1113
+ "agaric",
1114
+ "gyromitra",
1115
+ "stinkhorn, carrion fungus",
1116
+ "earthstar",
1117
+ "hen-of-the-woods, hen of the woods, Polyporus frondosus, Grifola frondosa",
1118
+ "bolete",
1119
+ "ear, spike, capitulum",
1120
+ "toilet tissue, toilet paper, bathroom tissue"
1121
+ ],
1122
+ "length_penalty": 1.0,
1123
+ "max_length": 20,
1124
+ "min_length": 0,
1125
+ "model_args": {
1126
+ "features_only": true,
1127
+ "in_chans": 3,
1128
+ "out_indices": [
1129
+ 0,
1130
+ 1,
1131
+ 2,
1132
+ 3
1133
+ ]
1134
+ },
1135
+ "model_type": "timm_wrapper",
1136
+ "no_repeat_ngram_size": 0,
1137
+ "num_beam_groups": 1,
1138
+ "num_beams": 1,
1139
+ "num_classes": 1000,
1140
+ "num_features": 384,
1141
+ "num_return_sequences": 1,
1142
+ "output_attentions": false,
1143
+ "output_hidden_states": false,
1144
+ "output_scores": false,
1145
+ "pad_token_id": null,
1146
+ "prefix": null,
1147
+ "pretrained_cfg": {
1148
+ "classifier": [
1149
+ "head.head.l",
1150
+ "head.head_dist.l"
1151
+ ],
1152
+ "crop_mode": "center",
1153
+ "crop_pct": 0.95,
1154
+ "custom_load": false,
1155
+ "first_conv": "stem.conv1.c",
1156
+ "fixed_input_size": false,
1157
+ "input_size": [
1158
+ 3,
1159
+ 224,
1160
+ 224
1161
+ ],
1162
+ "interpolation": "bicubic",
1163
+ "mean": [
1164
+ 0.485,
1165
+ 0.456,
1166
+ 0.406
1167
+ ],
1168
+ "pool_size": [
1169
+ 7,
1170
+ 7
1171
+ ],
1172
+ "std": [
1173
+ 0.229,
1174
+ 0.224,
1175
+ 0.225
1176
+ ],
1177
+ "tag": "dist_in1k"
1178
+ },
1179
+ "problem_type": null,
1180
+ "pruned_heads": {},
1181
+ "remove_invalid_values": false,
1182
+ "repetition_penalty": 1.0,
1183
+ "return_dict": true,
1184
+ "return_dict_in_generate": false,
1185
+ "sep_token_id": null,
1186
+ "suppress_tokens": null,
1187
+ "task_specific_params": null,
1188
+ "temperature": 1.0,
1189
+ "tie_encoder_decoder": false,
1190
+ "tie_word_embeddings": true,
1191
+ "tokenizer_class": null,
1192
+ "top_k": 50,
1193
+ "top_p": 1.0,
1194
+ "torchscript": false,
1195
+ "typical_p": 1.0
1196
+ },
1197
+ "backbone_feature_sizes": [
1198
+ [
1199
+ 256,
1200
+ 256
1201
+ ],
1202
+ [
1203
+ 128,
1204
+ 128
1205
+ ],
1206
+ [
1207
+ 64,
1208
+ 64
1209
+ ]
1210
+ ],
1211
+ "fpn_hidden_size": 256,
1212
+ "fpn_kernel_size": 1,
1213
+ "fpn_padding": 0,
1214
+ "fpn_stride": 1,
1215
+ "fpn_top_down_levels": [
1216
+ 2,
1217
+ 3
1218
+ ],
1219
+ "hidden_act": "gelu",
1220
+ "initializer_range": 0.02,
1221
+ "layer_norm_eps": 1e-06,
1222
+ "model_type": "edgetam_vision_model",
1223
+ "num_feature_levels": 3
1224
+ }
1225
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8858f8e4757b0b96dab8763f296ecffd845efbbbf698f64163cfa20a63d5fff4
3
+ size 55912824
preprocessor_config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": null,
3
+ "data_format": "channels_first",
4
+ "default_to_square": true,
5
+ "device": null,
6
+ "disable_grouping": null,
7
+ "do_center_crop": null,
8
+ "do_convert_rgb": true,
9
+ "do_normalize": true,
10
+ "do_pad": null,
11
+ "do_rescale": true,
12
+ "do_resize": true,
13
+ "image_mean": [
14
+ 0.485,
15
+ 0.456,
16
+ 0.406
17
+ ],
18
+ "image_processor_type": "Sam2ImageProcessorFast",
19
+ "image_std": [
20
+ 0.229,
21
+ 0.224,
22
+ 0.225
23
+ ],
24
+ "input_data_format": null,
25
+ "mask_size": {
26
+ "height": 256,
27
+ "width": 256
28
+ },
29
+ "pad_size": null,
30
+ "processor_class": "Sam2VideoProcessor",
31
+ "resample": 2,
32
+ "rescale_factor": 0.00392156862745098,
33
+ "return_tensors": null,
34
+ "size": {
35
+ "height": 1024,
36
+ "width": 1024
37
+ }
38
+ }
processor_config.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "point_pad_value": -10,
3
+ "processor_class": "Sam2VideoProcessor",
4
+ "target_size": 1024
5
+ }
video_preprocessor_config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": null,
3
+ "data_format": "channels_first",
4
+ "default_to_square": true,
5
+ "device": null,
6
+ "do_center_crop": null,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "do_sample_frames": null,
12
+ "fps": null,
13
+ "image_mean": [
14
+ 0.485,
15
+ 0.456,
16
+ 0.406
17
+ ],
18
+ "image_std": [
19
+ 0.229,
20
+ 0.224,
21
+ 0.225
22
+ ],
23
+ "input_data_format": null,
24
+ "num_frames": null,
25
+ "pad_size": null,
26
+ "processor_class": "Sam2VideoProcessor",
27
+ "resample": 2,
28
+ "rescale_factor": 0.00392156862745098,
29
+ "return_metadata": false,
30
+ "size": {
31
+ "height": 1024,
32
+ "width": 1024
33
+ },
34
+ "video_metadata": null,
35
+ "video_processor_type": "Sam2VideoVideoProcessor"
36
+ }