anymove

Animate any rigged creature from one AI video. The trick is the frame: one image holds two orthographic views of the creature at the same scale, front on the left and side on the right, sharing the vertical axis. A joint sits on the same pixel row in both halves, so the front view gives left-right and height, the side view gives front-back and height, and together they pin the joint in 3D. The method is @CanItCode's; the LoRA step is @Graalitoo's idea.

The pipeline

step	once per	what happens	skill
1. mesh	creature	a textured 3D model (Tripo H3.1 from one image, or your own GLB), facing -Y, real size	anymove-rig
2. skeleton	creature	joints placed in assets/<c>/rig.json, rig built and skinned in Blender	anymove-rig
3. limb colours	creature	each limb painted its own colour so left and right never swap	anymove-rig
4. two cameras	move	the painted reference frame plus a strict prompt go to MiniMax H3 Max on fal	anymove-film
5. teach a reader	creature	2,200 random poses rendered with known joints; DINOv2 learns this creature	anymove-reader
6. read every joint	clip	the reader finds every joint in both views, every frame	anymove-capture
7. two views, 3D	clip	shared heights tie the views; the pair gives each joint in 3D	anymove-capture
8. the rig follows	clip	forward kinematics in torch fitted to both views; GLB out	anymove-capture
9. the pack	creature	score the takes, keep the best, cut loops and root motion; one GLB with every move	anymove-pack
optional	creature	a MiniMax H3 LoRA that keeps the two-camera layout and colours more reliably	anymove-lora

Before anything

0. Check it's the right tool. If the person only wants a creature, look in Patch's Assets first: nine are ready. anymove suits big single gestures, and doesn't do eight-legged walking, lying down, bird wings or tail swipes (see the README). Check there's an NVIDIA card before anything else; without one, stop and say so.
1. Work in an anymove checkout. If the working directory has no tools/capture.sh, clone it: git clone https://github.com/Patch-gg/anymove && cd anymove.
2. Set up once: python3.11 -m venv .venv, then .venv/bin/pip install torch==2.11.0 --index-url https://download.pytorch.org/whl/cu128 and .venv/bin/pip install -r requirements.txt. Blender 5.x must be on PATH, or set BLENDER=/path/to/blender. On Windows, run everything in Git Bash, use py -3.11 -m venv .venv and .venv/Scripts/ where these say .venv/bin/, and set BLENDER. tools/*.sh handle the rest. If you call Blender yourself rather than through them, give it absolute paths: on Windows it reads a relative output path from the drive root.
3. For video generation and LoRA training, put FAL_KEY=... in .env (git ignores it) or export it. Never print it.
4. Every command takes a creature name; its config is assets/<c>/creature.json and every output lands in runs/<c>/ (reference frame, dataset, reader.pt, videos, clips). The scripts set MOCAP_CREATURE for you; when you run a python -m mocap... module by hand, set MOCAP_CREATURE=assets/<c> first, or the wolf's layout applies.

Which step next

• No assets/<c>/model.glb: make or fetch the mesh. Examples: tools/fetch_examples.sh <c> (wolf, dragon, spider, demon). New creature: anymove-rig.
• No assets/<c>/rigged/rigged.blend or runs/<c>/ref/ref.png: tools/prep.sh <c> then tools/rig.sh <c> (anymove-rig).
• No runs/<c>/reader.pt: anymove-reader (about 45 min on one RTX 4090), or tools/fetch_examples.sh <c> --readers.
• No clip for the move: anymove-film (about $0.26 a 5 s clip, and about 7 clips for a usable move).
• A clip: anymove-capture, tools/capture.sh <c> <video.mp4> <name>.
• Several good takes: anymove-pack, tools/pack.sh <c>.
• Clips keep swapping limb colours or the halves drift apart: anymove-lora.

Try it without a GPU

The repo carries the dragon's roar as read by its reader. Fit it on the CPU in about a minute and a half:

mkdir -p runs/dragon/clips/roar
MOCAP_CREATURE=assets/dragon .venv/bin/python -m mocap.solve examples/dragon_roar/det.npz \
  assets/dragon/rigged/rig_info.json runs/dragon/clips/roar/motion.json

It prints the paw agreement (about 0.93) and the reprojection error per view; motion.json holds the keyframes.

Costs and time (October 2026, one RTX 4090)

Quote the all-in numbers, not the price of one clip: about $3 for a usable move, and about $17 and a couple of hours for a new creature with about 5 moves (about $11 of AI services on the person's own card). A move took about 7 clips on average ($1.80), and 58% of captured clips passed the checks.

• Tripo H3.1 mesh with HD texture: $0.40. MiniMax H3 Max clip: about $0.26 for 5 s (H3: $0.30).
• Dataset and reader training: about 75 min on one RTX 4090. Reading a clip: about 12 s. Fitting: about 45 s. GLB and renders: about 3 min.
• A LoRA: $0.01 a training step, so 1,200 steps cost about $12 and take about 2 h on fal.

Say what a step will cost before calling fal, and confirm with the person before spending on a batch.