Upload folder using huggingface_hub

README.md

@@ -0,0 +1,38 @@
+---
+license: apache-2.0
+tags:
+  - mlx
+  - depth-estimation
+  - video
+  - video-depth-anything
+---
+
+# Video Depth Anything Small (MLX)
+
+MLX conversion of [depth-anything/Video-Depth-Anything-Small](https://huggingface.co/depth-anything/Video-Depth-Anything-Small) (ByteDance, CVPR 2025 highlight). Consistent monocular video depth estimation: DINOv2-vits backbone + temporal DPT head.
+
+Converted with `python -m mlx_vlm.models.video_depth_anything.convert --variant small`.
+
+## Usage
+
+```python
+from pathlib import Path
+from mlx_vlm.utils import load_model, get_model_path
+from mlx_vlm.models.video_depth_anything.processing_video_depth_anything import (
+    VideoDepthProcessor,
+)
+from mlx_vlm.models.video_depth_anything.generate import (
+    VideoDepthPredictor,
+    read_video_frames,
+)
+
+path = get_model_path("jarjoura/video-depth-anything-small-mlx")
+model = load_model(path)
+processor = VideoDepthProcessor.from_pretrained(path)
+predictor = VideoDepthPredictor(model, processor)
+
+frames, fps = read_video_frames("input.mp4", max_len=300, target_fps=15)
+depths = predictor.infer(frames)  # (T, H, W) float32 depth maps
+```
+
+Validated against the PyTorch reference: ~1e-5 max relative error on CPU, ~1% on GPU (Metal fast-math).

config.json

@@ -0,0 +1,14 @@
+{
+  "model_type": "video_depth_anything",
+  "encoder": "vits",
+  "metric": false,
+  "features": 64,
+  "out_channels": [
+    48,
+    96,
+    192,
+    384
+  ],
+  "num_frames": 32,
+  "pe": "ape"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5e5d669740f1f3031fbef66302aee61c50db27fe4035f9599e2b9eb09f7fea1
+size 116362301

preprocessor_config.json

@@ -0,0 +1,3 @@
+{
+  "input_size": 518
+}