Communitygithub.com

TerminalSkills/skills

Automatisez les workflows de capture de mouvement et de suivi dans Blender avec Python. Utilisez cette compétence lorsque l'utilisateur souhaite importer des données mocap BVH ou FBX, retargeter du mouvement sur des armatures, suivre le mouvement de caméra ou d'objet à partir de vidéo, résoudre le mouvement de caméra, nettoyer des données de capture de mouvement ou scripter n'importe quel pipeline de suivi dans Blender.

Qu'est-ce que skills ?

Cette compétence permet d'automatiser intégralement les pipelines de capture de mouvement (mocap) et de suivi vidéo dans Blender via des scripts Python exécutés depuis le terminal. Elle prend en charge l'import de fichiers BVH et FBX, le retargeting d'animations entre différentes armatures, le suivi de mouvement de caméra et d'objets à partir de séquences vidéo, la résolution 3D de trajectoires de caméra, ainsi que le nettoyage et la décimation des données d'animation. Les fonctionnalités incluent également l'export des données traitées, le baking de contraintes en keyframes, et l'exécution batch sur des bibliothèques de mocap. C'est un outil essentiel pour les techniciens de suivi, animateurs et développeurs travaillant avec Blender en mode headless ou automatisé.

Compatible avec~Claude Code~Codex CLI~Cursor
npx skills add https://github.com/TerminalSkills/skills/tree/HEAD/skills/blender-motion-capture

Demander à votre IA préférée

Ouvre une nouvelle conversation avec cette compétence d'agent déjà préchargée.

Documentation

Capture de mouvement Blender

Présentation

Importez, traitez et retargetez des données de capture de mouvement dans Blender avec Python. Travaillez avec des fichiers mocap BVH/FBX, suivez les mouvements de caméra et d'objets à partir de séquences vidéo, résolvez des trajectoires 3D de caméra et nettoyez des données d'animation — le tout scriptable depuis le terminal.

Instructions

1. Importer des fichiers BVH de capture de mouvement

import bpy

bpy.ops.import_anim.bvh(
    filepath="/chemin/vers/mocap.bvh",
    target='ARMATURE',
    global_scale=1.0,
    frame_start=1,
    use_fps_scale=False,
    rotate_mode='NATIVE',
    axis_forward='-Z', axis_up='Y'
)

armature = bpy.context.active_object
action = armature.animation_data.action
print(f"Importé : {armature.name}, Os : {len(armature.data.bones)}, Images : {action.frame_range}")

2. Importer FBX avec animation

bpy.ops.import_scene.fbx(
    filepath="/chemin/vers/mocap.fbx",
    use_anim=True,
    ignore_leaf_bones=True,
    automatic_bone_orientation=True,
    primary_bone_axis='Y', secondary_bone_axis='X'
)

3. Retargeter le mouvement entre armatures

from mathutils import Matrix

def retarget_motion(source_armature, target_armature, bone_mapping):
    """Retargetter l'animation en utilisant un mapping de noms d'os : {os_cible: os_source}"""
    source_action = source_armature.animation_data.action
    frame_start, frame_end = int(source_action.frame_range[0]), int(source_action.frame_range[1])

    if not target_armature.animation_data:
        target_armature.animation_data_create()
    new_action = bpy.data.actions.new(f"{source_action.name}_retarget")
    target_armature.animation_data.action = new_action

    for frame in range(frame_start, frame_end + 1):
        bpy.context.scene.frame_set(frame)
        for tgt_name, src_name in bone_mapping.items():
            src = source_armature.pose.bones.get(src_name)
            tgt = target_armature.pose.bones.get(tgt_name)
            if not src or not tgt:
                continue
            tgt.rotation_quaternion = src.rotation_quaternion
            tgt.keyframe_insert(data_path="rotation_quaternion", frame=frame)
            # Copier la position uniquement pour l'os racine
            if src_name == list(bone_mapping.values())[0]:
                tgt.location = src.location
                tgt.keyframe_insert(data_path="location", frame=frame)

# Exemple de mapping Mixamo → Rigify
mapping = {
    "spine": "mixamorig:Hips", "spine.001": "mixamorig:Spine",
    "spine.004": "mixamorig:Neck", "spine.006": "mixamorig:Head",
    "upper_arm.L": "mixamorig:LeftArm", "forearm.L": "mixamorig:LeftForeArm",
    "upper_arm.R": "mixamorig:RightArm", "forearm.R": "mixamorig:RightForeArm",
    "thigh.L": "mixamorig:LeftUpLeg", "shin.L": "mixamorig:LeftLeg",
    "thigh.R": "mixamorig:RightUpLeg", "shin.R": "mixamorig:RightLeg",
}

4. Nettoyer les données de capture de mouvement

def decimate_fcurve(fcurve, factor=0.5):
    """Supprimer des keyframes pour réduire les données tout en conservant la forme."""
    points = fcurve.keyframe_points
    total = len(points)
    keep_every = max(1, int(1.0 / factor))
    remove_indices = [i for i in range(total) if i % keep_every != 0 and i != 0 and i != total - 1]
    for i in reversed(remove_indices):
        points.remove(points[i])

armature = bpy.context.active_object
action = armature.animation_data.action
for fcurve in action.fcurves:
    decimate_fcurve(fcurve, factor=0.5)
    fcurve.update()

5. Suivi de mouvement vidéo et résolution de caméra

# Charger la séquence
clip = bpy.data.movieclips.load("/chemin/vers/footage.mp4")
scene = bpy.context.scene
scene.active_clip = clip

# Configurer le suivi
tracking = clip.tracking
settings = tracking.settings
settings.default_pattern_size = 21
settings.default_search_size = 71
settings.default_motion_model = 'AFFINE'

# Paramètres de la caméra pour la résolution
camera = tracking.camera
camera.sensor_width = 36.0
camera.focal_length = 50.0

# Résoudre le mouvement de la caméra
bpy.ops.clip.solve_camera()
solve_error = tracking.reconstruction.average_error
print(f"Erreur de résolution : {solve_error:.4f} px ({'Bon' if solve_error < 0.5 else 'Nécessite un raffinement'})")

# Configurer la scène à partir des données résolues
bpy.ops.clip.setup_tracking_scene()

6. Appliquer le mouvement suivi aux objets

obj = bpy.data.objects["MonObjet"]
constraint = obj.constraints.new(type='FOLLOW_TRACK')
constraint.clip = clip
constraint.track = tracking.tracks["Marker_01"]
constraint.use_3d_position = True
constraint.camera = scene.camera

# Baker la contrainte en keyframes
bpy.context.view_layer.objects.active = obj
obj.select_set(True)
bpy.ops.nla.bake(
    frame_start=1, frame_end=clip.frame_duration,
    only_selected=True, visual_keying=True,
    clear_constraints=True, bake_types={'OBJECT'}
)

7. Exporter les données d'animation

# Exporter en BVH
bpy.ops.export_anim.bvh(
    filepath="/tmp/output_mocap.bvh",
    frame_start=int(action.frame_range[0]),
    frame_end=int(action.frame_range[1]),
    rotate_mode='NATIVE'
)

# Exporter en FBX avec animation bakée
bpy.ops.export_scene.fbx(
    filepath="/tmp/output_anim.fbx",
    use_selection=True, bake_anim=True,
    bake_anim_use_all_bones=True, add_leaf_bones=False
)

Exemples

Exemple 1 : Parcourir une bibliothèque mocap en batch

Demande utilisateur : "Importer tous les fichiers BVH d'un dossier, lister le nombre d'os et les plages d'images"

import bpy, glob, os

for filepath in sorted(glob.glob("/chemin/vers/bibliotheque_mocap/*.bvh")):
    bpy.ops.object.select_all(action='SELECT')
    bpy.ops.object.delete()
    bpy.ops.import_anim.bvh(filepath=filepath, target='ARMATURE', global_scale=0.01, frame_start=1)
    arm = bpy.context.active_object
    if arm and arm.animation_data:
        action = arm.animation_data.action
        duree = (action.frame_range[1] - action.frame_range[0]) / bpy.context.scene.render.fps
        print(f"{os.path.basename(filepath)} : {len(arm.data.bones)} os, {duree:.1f}s")

Exécutez : blender --background --python scan_mocap.py

Exemple 2 : Appliquer la mocap à un personnage et effectuer le rendu

Demande utilisateur : "Importer un fichier BVH, l'appliquer à mon personnage riggé, et effectuer le rendu d'un aperçu"

import bpy

bpy.ops.wm.open_mainfile(filepath="/chemin/vers/personnage.blend")
char_armature = bpy.data.objects["Armature"]

bpy.ops.import_anim.bvh(filepath="/chemin/vers/walk_cycle.bvh", target='ARMATURE', global_scale=0.01)
mocap_armature = bpy.context.active_object
mocap_action = mocap_armature.animation_data.action

# Transférer l'action (fonctionne lorsque les noms d'os correspondent)
if not char_armature.animation_data:
    char_armature.animation_data_create()
char_armature.animation_data.action = mocap_action

# Supprimer l'armature temporaire, définir la plage d'images, ajouter une caméra, effectuer le rendu
bpy.data.objects.remove(mocap_armature)
scene = bpy.context.scene
scene.frame_start, scene.frame_end = int(mocap_action.frame_range[0]), int(mocap_action.frame_range[1])
scene.render.filepath = "/tmp/mocap_preview/image_"
bpy.ops.render.render(animation=True)

Directives

  • Le BVH est le plus simple (texte clair avec hiérarchie + mouvement). FBX prend en charge des données plus riches (blend shapes, plusieurs prises).
  • L'échelle compte : les fichiers BVH utilisent souvent les centimètres. Définissez global_scale=0.01 pour les fichiers basés en cm.
  • La correspondance des noms d'os est essentielle pour le retargeting. Construisez un dictionnaire de mapping pour chaque format source.
  • Pour le retargeting, copiez les rotations pour tous les os mais uniquement la position pour l'os racine/hanche.
  • Nettoyez la mocap importée en décimant les keyframes — la mocap brute a des keyframes à chaque image, ce qui rend l'édition difficile.
  • La qualité de la résolution de caméra dépend du nombre et de la répartition des marqueurs. Utilisez 8+ marqueurs bien répartis, maintenez l'erreur en dessous de 0,5 px.
  • Utilisez bpy.ops.nla.bake() pour convertir les contraintes en keyframes pour l'export.
  • Exportez toujours avec bake_anim=True dans FBX pour aplatir les bandes NLA et les contraintes.

Individual skills in this repo

This repo contains 20 individual skills — each has its own dedicated page.

TerminalSkills/skills

>- Run GitHub Actions locally with act. Use when a user asks to test GitHub Actions workflows locally, debug CI pipelines without pushing, or run workflows offline.

TerminalSkills/skills

>- You are an expert in AG2 (formerly AutoGen), the open-source multi-agent conversation framework. You help developers build systems where multiple AI agents collaborate through structured conversations — with tool use, human-in-the-loop, code execution, group chat orchestration, and nested conversations — for complex tasks like software development, research, and data analysis.

TerminalSkills/skills

>- Generate short-form videos with AI — script writing, text-to-speech narration, stock footage selection, subtitle generation, and video assembly. Use when: creating TikTok/YouTube Shorts/Reels content, automating video production, building content pipelines.

TerminalSkills/skills

>- Animate 3D objects and characters in Blender with Python. Use when the user wants to keyframe properties, create armatures and rigs, set up IK/FK chains, animate shape keys for facial animation, edit F-Curves, use the NLA editor to blend actions, add drivers for expression-based animation, or script any animation workflow in Blender.

TerminalSkills/skills

>- You are an expert in Bolt.new by StackBlitz, the AI-powered full-stack development environment that runs entirely in the browser. You help developers go from idea to deployed app in minutes using natural language prompts — Bolt generates complete applications with frontend, backend, database, and deployment, all running in a WebContainer without local setup.

TerminalSkills/skills

>- Assists with using Bun as an all-in-one JavaScript/TypeScript runtime, package manager, bundler, and test runner. Use when building HTTP servers, managing packages, running tests, or migrating from Node.js. Trigger words: bun, bun serve, bun install, bun test, bun build, javascript runtime, bun runtime.

TerminalSkills/skills

>- Self-host customer support with Chatwoot. Use when a user asks to set up open-source customer support, add live chat without SaaS costs, build a multi-channel inbox, or deploy a free Intercom alternative.

TerminalSkills/skills

>- You are an expert in Chi, the lightweight, idiomatic Go HTTP router built on `net/http`. You help developers build composable HTTP services using Chi's middleware stack, route groups, URL parameters, sub-routers, and context-based request scoping — providing Express-like ergonomics while staying 100% compatible with Go's standard library.

TerminalSkills/skills

When the user wants to edit, review, or improve existing marketing copy. Also use when the user mentions 'edit this copy,' 'review my copy,' 'copy feedback,' 'proofread,' 'polish this,' 'make this better,' or 'copy sweep.' This skill provides a systematic approach to editing marketing copy through multiple focused passes.

TerminalSkills/skills

>- Add live chat and customer support with Crisp. Use when a user asks to add a chat widget, implement live customer support, set up a help desk, create a knowledge base, or add a chatbot to a website.

TerminalSkills/skills

>- Assists with building custom interactive data visualizations using D3.js. Use when creating charts, graphs, maps, force layouts, or hierarchical diagrams that require fine-grained control beyond what charting libraries provide. Trigger words: d3, data visualization, chart, svg, scales, force graph, treemap, choropleth.

TerminalSkills/skills

dbt (data build tool) transforms data in your warehouse using SQL SELECT statements. Learn project setup, models, tests, documentation, incremental materializations, and integration with data warehouses like PostgreSQL, BigQuery, and Snowflake.

TerminalSkills/skills

>- You are an expert in dlt, the open-source Python library for building data pipelines. You help developers load data from any API, file, or database into warehouses and lakes using simple Python decorators — with automatic schema inference, incremental loading, and built-in data contracts. dlt is the "requests library for data pipelines.

TerminalSkills/skills

Data Version Control for ML projects. Track large datasets and models alongside Git, build reproducible ML pipelines, and run experiments with metric comparison. Works with any storage backend including S3, GCS, Azure, and local filesystems.

TerminalSkills/skills

>- You are an expert in E2B, the cloud platform for running AI-generated code in secure sandboxes. You help developers give AI agents the ability to execute code, install packages, read/write files, and run long processes in isolated cloud environments — each sandbox is a lightweight VM that boots in ~150ms with full Linux, filesystem, and networking.

TerminalSkills/skills

>- Transcode, convert, edit, and process audio and video with FFmpeg. Use when a user asks to convert video formats (webm to mp4, mkv to mp4), extract audio from video, compress video files, trim or cut clips, concatenate videos, add subtitles, create thumbnails, apply filters, change resolution or bitrate, re-encode media, create GIFs from video, add watermarks, normalize audio, stream media, or build automated media processing pipelines.

TerminalSkills/skills

>- Create advanced animations with GSAP. Use when building complex timelines, scroll-driven animations, SVG morphing, text animations, or high-performance web animations beyond CSS capabilities.

TerminalSkills/skills

>- Generate deterministic MP4 videos from HTML, CSS, media, and seekable animations using HeyGen's HyperFrames framework. Use when someone asks to "render HTML to video", "make a video with HyperFrames", "create a programmatic video", "turn an animation into MP4", "build a launch/product video from HTML", "render GSAP/Lottie/Three.js to video", or set up an agent-driven video pipeline. Covers init/preview/render/add/lint/inspect commands, data-* timing attributes, animation adapters, and the component catalog.

TerminalSkills/skills

When the user wants to perform load testing, stress testing, or performance testing of APIs and websites using k6. Also use when the user mentions "k6," "load test," "performance test," "stress test," "spike test," "soak test," "thresholds," "virtual users," or "VUs." For browser-based testing, see selenium.

TerminalSkills/skills

>- Lovable (formerly GPT Engineer) is a browser-based AI app builder that turns a natural-language description into a working full-stack web app with a built-in backend, authentication, database and one-click publishing. Use when the user wants to prototype or ship a web app with Lovable, write effective Lovable prompts, connect Supabase or GitHub, review generated security rules, or move a Lovable project into their own editor.

Skills associés