GPU: lower requested ZeroGPU duration 180 -> 120s (free-tier limit)

Face crop: square again (better for the model), but tight to face+hair

Face crop: tighter, non-square box around face+hair (less background)

Face crop: make it hair-aware (don't clip top hair; cover long hair)

Inputs: add default-on face-centric crop (pixel3dmm-style)

UI: reorder 2D video tabs to Normals, Depth, Canonical, Tracks

Viewer: start the camera closer to the point cloud (1.7x -> 1.15x fit)

Viewer: default the colorful-tracks toggle to OFF (show RGB by default)

Viewer: tracks/RGB toggle + restrict colorful tracks to face+hair (facer)

Viewer: preload all frames in a client-side three.js player (smooth playback)

Viewer: animate the point cloud over time (play/pause + speed)

Examples: show a NeRSemble thumbnail directly + bold dataset label

Default to one-by-one, robust video decode, add 40-img/1-img examples

Viewer: keep the full point cloud (no subsampling)

Demo feedback: video upload, fix washed-out points, smoother scrub, rename folder

Apply demo feedback: predicted poses, title, dual point clouds, working viewer

Add torchvision (not in the ZeroGPU base image)

Fix missing model deps on the Space (lazy export + addict/evo)

Fix 'No images found' on upload (robust input normalization)

Fix RVM background removal crash on the Space (trust_repo + fallback)

Fix Slider min==max crash on gradio 6.19

Docs: mount model repo as a read-only volume for the checkpoint

Pull checkpoint from a HF model repo at startup

Add FaceAnything Gradio demo app

initial commit