Skip to main content
skill CX Codex

anymove-reader — Claude Code skill: what it does, how to install

anymove-reader is a Claude Code skill that trains a DINOv2 joint detector for a 3D creature. It runs between anymove-rig setup and anymove-capture mocap tracking.

Apache-2.0 ★ 653 updated 2026-11-13

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add patch-gg/anymove
/plugin install anymove@anymove
Open on GitHub

What anymove-reader does

The skill renders roughly 2,200 synthetic poses in Blender 5.x from dual layout cameras and fine-tunes DINOv2 (ViT-B/14 with registers) along with a heatmap head. Training data includes motion blur, turns, random lighting, and video degradations like downscaling and JPEG noise. An optional exam script renders a known motion to score tracking error in centimetres.

Execution requires an NVIDIA GPU with CUDA (around 12 GB VRAM at batch 20, ideally a 24 GB card), Blender 5.x, and a Python 3.11 virtual environment. The first run fetches facebook/dinov2-with-registers-base. Running training beside more than two Blender Cycles renders exhausts a 24 GB card under WSL.

What it helps with

  • Render 2,200 synthetic pose samples from Blender cameras
  • Fine-tune DINOv2 on creature joint heatmaps
  • Evaluate 3D tracking error in centimetres on a benchmark motion
  • Fix swapped or misplaced joints on processed video clips

Example prompts for the agent

  • Render the pose dataset for creature wolf and start anymove-reader training
  • Run the exam motion test to score joint error on runs/dragon
  • Generate 2200 training poses for creature demon using tools/dataset.sh

How to install anymove-reader

1Plugin marketplace

/plugin marketplace add patch-gg/anymove
/plugin install anymove@anymove

This installs the whole anymove plugin with all its skills

2Skills CLI (Vercel)

npx skills add patch-gg/anymove --skill anymove-reader -a claude-code

3Manually: put the files in place

~/.claude/skills/anymove-reader/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/anymove-reader/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about anymove-reader

How to install anymove-reader in Claude Code?

Run /plugin marketplace add patch-gg/anymove and then execute /plugin install anymove@anymove.

What does anymove-reader do?

It renders multi-view poses from Blender rigs and fine-tunes a vision transformer to track creature joints.

What is required to run anymove-reader?

An NVIDIA GPU with CUDA (~12 GB VRAM for batch 20), Blender 5.x, and a Python 3.11 virtual environment are required.

Source text

SKILL.md · Apache-2.0
---
name: anymove-reader
description: Train the joint reader for one anymove creature - render about 2,200 random poses from both layout cameras with every joint's true position, fine-tune DINOv2 (ViT-B/14 with registers) plus a heatmap head on them, check the held-out error, and optionally run the exam (a made-up motion with known joints, read back and scored in centimetres). Use after anymove-rig and before anymove-capture, when a reader misses or swaps joints on real clips, or when planning GPU time for dataset rendering and training.
license: Apache-2.0
compatibility: NVIDIA GPU with CUDA (about 12 GB at batch 20, so a 24 GB card), Blender 5.x, Python 3.11 venv from requirements.txt. Downloads facebook/dinov2-with-registers-base on first use.
---

# anymove-reader: teach DINOv2 one creature

The reader is trained on the creature itself, so it knows every joint name and every limb colour. Data is free:
Blender poses the rig at random inside its joint limits and records where each joint lands in both views.

## 1. Render the dataset (about 25 min for 2,200 poses on one RTX 4090)

```bash
tools/dataset.sh <c> 2200 3     # 3 Blender processes -> runs/<c>/dataset/{00000_front.png, _side.png, _meta.json}
```

Half the poses spread uniformly over the limits, half stay near rest; the pelvis sometimes rears; half carry motion
blur from a random velocity (video models smear fast limbs); lights, turn (±30°, sometimes ±70°), position and
airtime are random. Open `runs/<c>/dataset_samples.jpg`: dots must sit on the joints. Interrupted renders resume:
samples that exist are skipped.

GPU budget on a 24 GB card under WSL: a reader training (about 12 GB) plus more than two Blender Cycles renders fills
the card and everything stalls. Render first, or train beside at most two renders. Blender uses CUDA (OptiX is
unreliable under WSL).

## 2. Train (17-20 min, 20 epochs)

```bash
tools/train.sh <c> 20 20        # runs/<c>/reader.pt; prints held-out error per epoch in layout pixels
```

Every render is laid on a fresh random backdrop and degraded like video (framing jitter, colour, blur, downscale,
noise, JPEG) so the reader works on AI clips, not just clean renders. 4% of the samples are held out. The checkpoint
is rewritten after each epoch.

What good looks like (front / side, layout pixels, after 20 epochs): spider 2.9 / 3.6, dragon 6.2 / 5.5, wolf
6.9 / 3.8, demon 8.4 / 9.4. Judge it against the creature's size in the frame: the demon stands 412 px tall, so
8.4 px is about 2% of its height, like the dragon. If the error stalls well above that, check the dataset sheet (dots
off the joints means the layout in creature.json changed after rendering) and the joint limits.

## 3. The exam (optional, about 10 min): known motion in, error in centimetres out

A random smooth motion inside the limits is rendered in the same studio, read back and scored per joint.

```bash
export MOCAP_CREATURE=assets/<c>; R=assets/<c>/rigged; E=runs/<c>/exam
.venv/bin/python -m mocap.exam make $R/rig_info.json $E/truth.json 0 120
blender -b -P blender/apply_motion.py -- $R/rigged.blend $R/rig_info.json $E/truth.json $E/truth --views front,side --paint --blur
.venv/bin/python -m mocap.exam video $E/truth $E/exam.mp4 24
.venv/bin/python -m mocap.detect $E/exam.mp4 runs/<c>/reader.pt runs/<c>/ref/ref.png $E/det.npz --no-calib
.venv/bin/python -m mocap.solve $E/det.npz $R/rig_info.json $E/solved.json
blender -b -P blender/apply_motion.py -- $R/rigged.blend $R/rig_info.json $E/solved.json $E/solved
.venv/bin/python -m mocap.exam score $E/truth/joints.json $E/solved/joints.json
```

The wolf scored 0.94 cm mean 3D joint error on a 1.6 m body (median 0.63 cm, paws 1.5 cm, worst joint 3.4 cm).

## When the reader struggles on real clips

- Joints jump between limbs: the limb colours in the clip drifted from the paint (anymove-film), or two limbs share
  similar colours. Repaint far-apart colours and retrain.
- A pose the clip shows never appeared in training (lying on its side, a full curl): widen that bone's limits in
  rig.json, rebuild the rig, render and train again. Random poses don't cover lying-down; captured motions would.
- Fur, cloth or ears flapping: the wolf's tracked-fur reader added a tip joint per fur bone
  (`blender/add_fur_bones.py`, `blender/export_fur_rig_info.py`) and read swings of 7-28° from the video.

© Patch · Apache-2.0 · patch-gg/anymove

Similar skills

All skills for AI and ML →

From the same set7

Project sponsors

Companies that keep this analytics open