KevinMathew commited on
Commit
07ff927
·
1 Parent(s): 4cc8033

Add unofficial DynaDUSt3R .pth weights + model card

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ *.pth filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # DynaDUSt3R (Unofficial) — Model Weights
2
+
3
+ This repository hosts PyTorch `.pth` checkpoint files for the **DynaDUSt3R** model — an external, unofficial reimplementation of the model proposed in the paper [*Stereo4D: Learning How Things Move in 3D from Internet Stereo Videos*](https://arxiv.org/abs/2412.09621) by Linyi Jin, Richard Tucker, Zhengqi Li, David Fouhey, Noah Snavely, and Aleksander Hołyński.
4
+
5
+ ---
6
+
7
+ ## Overview
8
+
9
+ DynaDUSt3R extends [DUSt3R](https://openaccess.thecvf.com/content/CVPR2024/papers/Wang_DUSt3R_Geometric_3D_Vision_Made_Easy_CVPR_2024_paper.pdf) to dynamic scenes by predicting per-pixel 3D points and 3D motion trajectories from stereo video sequences.
10
+ These weights were obtained from an independent reimplementation and are provided solely for research and reproducibility purposes.
11
+
12
+ The accompanying codebase can be found in the [dynadust3r-unofficial](https://github.com/KevinMathewT/dynadust3r-unofficial) repository.
13
+ Instructions for running inference are available in the [inference example script](https://github.com/KevinMathewT/dynadust3r-unofficial/blob/main/infer/infer_example.py).
14
+
15
+ ---
16
+
17
+ ## Contents
18
+
19
+ - `.pth` checkpoint files corresponding to trained DynaDUSt3R models.
20
+ - Optional `SHA256SUMS.txt` for integrity verification.
21
+
22
+ These checkpoints are compatible with the model definitions in the [dynadust3r-unofficial](https://github.com/KevinMathewT/dynadust3r-unofficial) repository.
23
+
24
+ ---
25
+
26
+ ## Usage
27
+
28
+ Please follow the setup and inference steps described in the main [dynadust3r-unofficial](https://github.com/KevinMathewT/dynadust3r-unofficial) repository.
29
+ The weights can be loaded using `torch.load()` as shown in the [inference example](https://github.com/KevinMathewT/dynadust3r-unofficial/blob/main/infer/infer_example.py).
30
+
31
+ **Note:** `.pth` files use Python pickle and should only be loaded from trusted sources.
32
+
33
+ ---
34
+
35
+ ## References
36
+
37
+ - [Stereo4D Project Page](https://stereo4d.github.io/)
38
+ - [DUSt3R Paper (CVPR 2024)](https://openaccess.thecvf.com/content/CVPR2024/papers/Wang_DUSt3R_Geometric_3D_Vision_Made_Easy_CVPR_2024_paper.pdf)
39
+
40
+ ---
41
+
42
+ ## License
43
+
44
+ These weights are provided for research and non-commercial use only.
45
+ This repository is **unofficial** and not affiliated with the authors of *Stereo4D*.
SHA256SUMS.txt ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ c77856b375739c2dff48b29b14435ead8b50610bcb1afc6d6520f8c71cbcff99 best_loss_0.367080_iter_34000_epoch_17.pth
2
+ 690bbad58dc5d72dc2514ce9153fe407a1cc23c7a2c581bb508e8b5d0abe12a4 best_loss_0.386052_iter_38000_epoch_19.pth
3
+ ca5e42a1e84633223d41e885567804643f98bb7b1efd3090823fdd61947e2f29 best_loss_0.392713_iter_94000_epoch_47.pth
4
+ 80a68771cc5b9f33c261d62d1d3ff4b1ba0762212563daa673a9076a682aa647 best_loss_0.396031_iter_32000_epoch_16.pth
5
+ 7d53d590bf965d31dbf2b2c72e47492155ff344d6da197e146f5249eb737448a best_loss_0.427555_iter_24000_epoch_12.pth
6
+ 802be9dbcc3e6640508c66783cbfb925a27ad065c57bb83a8a433d2abc0dad10 best_loss_0.427857_iter_92000_epoch_46.pth
7
+ f57831485f74d5d400fd1b4ba9af79b8d48697d9e6daab90bc64164329a4c4c6 best_loss_0.431392_iter_52000_epoch_26.pth
8
+ f556c34852b859492e93df4805c334c4b9eab8c6c66c34475284818d3336e342 best_loss_0.443512_iter_88000_epoch_44.pth
9
+ ff85a2967f150d8cf0207fe0ec3cdc56f0ff1d74eb7a5f2c456cc544be3f9548 best_loss_0.468565_iter_58000_epoch_29.pth
10
+ 1d04f49b509e2315f97338802eefd6908b09410d21cb31943257613235203a05 best_loss_0.478208_iter_56000_epoch_28.pth
11
+ 836ef966b12c4e5538202b9f6d3641b8c1a84878002d676faee5c5fd3083080b best_loss_0.480609_iter_40000_epoch_20.pth
12
+ 5e5c699bedc3f977dda657d46a84ed9022e82a087f3f822a60a7ba1a767b5cf7 best_loss_0.480798_iter_90000_epoch_45.pth
13
+ 8d81c961a3d7501b45411983f3cf020cb7362ad7123509f96413cf191272d540 best_loss_0.482799_iter_86000_epoch_43.pth
14
+ 2e72e9079f4813eda4dceb4a1ab895391f537ccab153ef5708adaabec2b0a52f best_loss_0.493017_iter_66000_epoch_33.pth
15
+ 52b2773f1a7a2bd37be5e1830e59c71610b7400200cf7751c7070f5b1750d7be best_loss_0.499884_iter_26000_epoch_13.pth
16
+ 1ae311f280b0a6787b06e35186bd850267a32556fe7779646b3f82681f469faa best_loss_0.503958_iter_82000_epoch_41.pth
17
+ cfe7d3a445b8fcd0caeac001df4b28abb81dafd6640a7efdc2fba2e40a454971 best_loss_0.523247_iter_42000_epoch_21.pth
18
+ ca89bb127294e4980c1ec7beca24944f997298684cf61797fcf5576566b08ab1 best_loss_0.525647_iter_28000_epoch_14.pth
19
+ a5efb50b3cf5bd19bc7bfc37040589c7697108a611f2bc2b877d78a7e7b40a04 best_loss_0.527959_iter_30000_epoch_15.pth
20
+ 6334b13c6089edbbd1cc4137ae58cab44dcc93d0e88c42d94a64e8991f3f32c9 best_loss_0.528999_iter_60000_epoch_30.pth
21
+ 4bd0414a5237631babc39247989a76da1a786d34bc47b8130e932f3da4f973d9 last_iter_98000_epoch_49.pth
best_loss_0.367080_iter_34000_epoch_17.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c77856b375739c2dff48b29b14435ead8b50610bcb1afc6d6520f8c71cbcff99
3
+ size 7315402520
best_loss_0.386052_iter_38000_epoch_19.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:690bbad58dc5d72dc2514ce9153fe407a1cc23c7a2c581bb508e8b5d0abe12a4
3
+ size 7315402520
best_loss_0.392713_iter_94000_epoch_47.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca5e42a1e84633223d41e885567804643f98bb7b1efd3090823fdd61947e2f29
3
+ size 7315402520
best_loss_0.396031_iter_32000_epoch_16.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:80a68771cc5b9f33c261d62d1d3ff4b1ba0762212563daa673a9076a682aa647
3
+ size 7315402520
best_loss_0.427555_iter_24000_epoch_12.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d53d590bf965d31dbf2b2c72e47492155ff344d6da197e146f5249eb737448a
3
+ size 7315402520
best_loss_0.427857_iter_92000_epoch_46.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:802be9dbcc3e6640508c66783cbfb925a27ad065c57bb83a8a433d2abc0dad10
3
+ size 7315402520
best_loss_0.431392_iter_52000_epoch_26.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f57831485f74d5d400fd1b4ba9af79b8d48697d9e6daab90bc64164329a4c4c6
3
+ size 7315402520
best_loss_0.443512_iter_88000_epoch_44.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f556c34852b859492e93df4805c334c4b9eab8c6c66c34475284818d3336e342
3
+ size 7315402520
best_loss_0.468565_iter_58000_epoch_29.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff85a2967f150d8cf0207fe0ec3cdc56f0ff1d74eb7a5f2c456cc544be3f9548
3
+ size 7315402520
best_loss_0.478208_iter_56000_epoch_28.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d04f49b509e2315f97338802eefd6908b09410d21cb31943257613235203a05
3
+ size 7315402520
best_loss_0.480609_iter_40000_epoch_20.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:836ef966b12c4e5538202b9f6d3641b8c1a84878002d676faee5c5fd3083080b
3
+ size 7315402520
best_loss_0.480798_iter_90000_epoch_45.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e5c699bedc3f977dda657d46a84ed9022e82a087f3f822a60a7ba1a767b5cf7
3
+ size 7315402520
best_loss_0.482799_iter_86000_epoch_43.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d81c961a3d7501b45411983f3cf020cb7362ad7123509f96413cf191272d540
3
+ size 7315402520
best_loss_0.493017_iter_66000_epoch_33.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e72e9079f4813eda4dceb4a1ab895391f537ccab153ef5708adaabec2b0a52f
3
+ size 7315402520
best_loss_0.499884_iter_26000_epoch_13.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52b2773f1a7a2bd37be5e1830e59c71610b7400200cf7751c7070f5b1750d7be
3
+ size 7315402520
best_loss_0.503958_iter_82000_epoch_41.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ae311f280b0a6787b06e35186bd850267a32556fe7779646b3f82681f469faa
3
+ size 7315402520
best_loss_0.523247_iter_42000_epoch_21.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfe7d3a445b8fcd0caeac001df4b28abb81dafd6640a7efdc2fba2e40a454971
3
+ size 7315402520
best_loss_0.525647_iter_28000_epoch_14.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca89bb127294e4980c1ec7beca24944f997298684cf61797fcf5576566b08ab1
3
+ size 7315402520
best_loss_0.527959_iter_30000_epoch_15.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a5efb50b3cf5bd19bc7bfc37040589c7697108a611f2bc2b877d78a7e7b40a04
3
+ size 7315402520
best_loss_0.528999_iter_60000_epoch_30.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6334b13c6089edbbd1cc4137ae58cab44dcc93d0e88c42d94a64e8991f3f32c9
3
+ size 7315402520
last_iter_98000_epoch_49.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4bd0414a5237631babc39247989a76da1a786d34bc47b8130e932f3da4f973d9
3
+ size 7315337842