π AnyTraverse Studio β Live Evaluation & HITL Dashboard
A live Gradio dashboard for evaluating the AnyTraverse zero-shot off-road
traversability framework (paper,
PyPI).
Workflow
- Upload a video of an off-road scene.
- Set the traversability preferences (Ο), scene-similarity threshold,
ROI-uncertainty threshold, ROI bounds and the VLM inference frame skip.
- Press βΆοΈ Go / Reset.
- Watch per frame:
- raw image + ROI box Β· traversability map Β· uncertainty map Β· ROI crop
- attention maps for all prompts (live, not in the exported video)
- live dual-metric ROI traversability + uncertainty line plot
- horizontal 0β1 gauge bars for the two ROI scores
- the traversal state (
ok / unknown_scene / unknown_object)
- The run halts whenever
traversal_state != OK. Provide an operator
update like mud: -0.7; gravel: 0.6 (or just ok) and press Resume.
Thresholds and ROI bounds can be edited live during a run β they are
applied directly to the running pipeline.
Notes
- The VLM weights (CLIPSeg + CLIP) download on first run and are cached.
- The composed analysis video is exported as an H.264
.mp4 (bundled
imageio-ffmpeg, no system ffmpeg required).
- GPU is recommended; the space is configured for a T4 accelerator.