Motif-Technologies
/

Motif-Vision-Encoder

Image Feature Extraction

feature-extraction

vision-transformer

self-supervised

Model card Files Files and versions

taewhan commited on Jul 23

Commit

8bc3bc2

·

verified ·

1 Parent(s): 8018605

Update README.md

Files changed (1) hide show

README.md +1 -2

README.md CHANGED Viewed

@@ -30,8 +30,7 @@ performance across image and video benchmarks — and leads on DAVIS video track
   (`base=100`), SwiGLU FFN, LayerScale, per-head QK-norm, gated attention, 4 register tokens.
 - **Tokenizer**: `Conv3d(kernel=stride=(tubelet, patch, patch))` — image `(B,3,H,W)` → `T=1`,
   video `(B,T,3,H,W)`. Token layout `[CLS] + [register × 4] + [patch × N]`.
-- **This repo**: inference-only. Weights + self-contained `modeling_motif_vision_encoder.py` (loaded
-  via `trust_remote_code`). Training/eval code lives in the GitHub repo.
 ## Usage

   (`base=100`), SwiGLU FFN, LayerScale, per-head QK-norm, gated attention, 4 register tokens.
 - **Tokenizer**: `Conv3d(kernel=stride=(tubelet, patch, patch))` — image `(B,3,H,W)` → `T=1`,
   video `(B,T,3,H,W)`. Token layout `[CLS] + [register × 4] + [patch × N]`.
 ## Usage