LivePortrait

LivePortrait は、ディープラーニングに基づくポートレート駆動生成モデルで、静止したポートレートをリアルな表情や動作を持つ動的な動画効果に変換することができます。

モデルを手動でダウンロードして Raspberry Pi 5 にアップロードするか、以下のコマンドでモデルリポジトリを取得します。

ヒント

git lfs がインストールされていない場合は、事前に git lfs インストール手順を参照してください。

git clone https://huggingface.co/AXERA-TECH/LivePortrait

ファイル内容説明:

m5stack@raspberrypi:~/rsp/LivePortrait $ ls -lh
total 16K
drwxrwxr-x 3 m5stack m5stack 4.0K Aug 13 09:41 assets
-rw-rw-r-- 1 m5stack m5stack    0 Aug 13 09:41 config.json
drwxrwxr-x 5 m5stack m5stack 4.0K Aug 13 09:41 python
-rw-rw-r-- 1 m5stack m5stack 5.7K Aug 13 09:41 README.md

仮想環境を作成

python -m venv lvpr

仮想環境を有効化

source lvpr/bin/activate

依存パッケージをインストール

pip install https://github.com/AXERA-TECH/pyaxengine/releases/download/0.1.3.rc1/axengine-0.1.3-py3-none-any.whl
pip install -r python/requirements.txt
pip install requests tqdm

Image

実行：

python ./python/infer.py --source ./assets/examples/source/s0.jpg --driving ./assets/examples/driving/d8.jpg --models ./python/axmodels/ --output-dir ./axmodel_infer

実行結果：

(lvpr) m5stack@raspberrypi:~/rsp/LivePortrait $ python ./python/infer.py --source ./assets/examples/source/s0.jpg --driving ./assets/examples/driving/d8.jpg --models ./python/axmodels/ --output-dir ./axmodel_infer
[INFO] Available providers:  ['AXCLRTExecutionProvider']
[INFO] Using provider: AXCLRTExecutionProvider
[INFO] SOC Name: AX650N
[INFO] VNPU type: VNPUType.DISABLED
[INFO] Compiler version: 3.3 144960ad
[INFO] Using provider: AXCLRTExecutionProvider
[INFO] SOC Name: AX650N
[INFO] VNPU type: VNPUType.DISABLED
[INFO] Compiler version: 3.3 144960ad
[INFO] Using provider: AXCLRTExecutionProvider
[INFO] SOC Name: AX650N
[INFO] VNPU type: VNPUType.DISABLED
[INFO] Compiler version: 3.3 0f7260e8
[INFO] Using provider: AXCLRTExecutionProvider
[INFO] SOC Name: AX650N
[INFO] VNPU type: VNPUType.DISABLED
[INFO] Compiler version: 3.3 144960ad
FaceAnalysisDIY warmup time: 0.237s
LandmarkRunner warmup time: 0.183s
2025-08-13 10:04:42.914 | INFO     | __main__:main:727 - Start making driving motion template...
2025-08-13 10:04:43.782 | INFO     | __main__:main:747 - Prepared pasteback mask done.
2025-08-13 10:04:44.565 | INFO     | __main__:main:787 - The output of image-driven portrait animation is an image.
2025-08-13 10:04:50.566 | DEBUG    | __main__:warp_decode:647 - warp time: 5.997s
2025-08-13 10:04:50.902 | INFO     | __main__:main:881 - Animated image: ./axmodel_infer/s0--d8.jpg
2025-08-13 10:04:50.902 | INFO     | __main__:main:882 - Animated image with concat: ./axmodel_infer/s0--d8_concat.jpg
2025-08-13 10:04:50.920 | DEBUG    | __main__:<module>:894 - LivePortrait axmodel infer time: 18.068s 

元画像：

出力画像：

Video

実行：

python ./python/infer.py --source ./assets/examples/source/s0.jpg --driving ./assets/examples/driving/d0.mp4 --models ./python/axmodels/ --output-dir ./axmodel_infer

実行結果：

(lvpr) m5stack@raspberrypi:~/rsp/LivePortrait $ python ./python/infer.py --source ./assets/examples/source/s0.jpg --driving ./assets/examples/driving/d0.mp4 --models ./python/axmodels/ --output-dir ./axmodel_infer

[INFO] Available providers:  ['AXCLRTExecutionProvider']
[INFO] Using provider: AXCLRTExecutionProvider
[INFO] SOC Name: AX650N
[INFO] VNPU type: VNPUType.DISABLED
[INFO] Compiler version: 3.3 144960ad
[INFO] Using provider: AXCLRTExecutionProvider
[INFO] SOC Name: AX650N
[INFO] VNPU type: VNPUType.DISABLED
[INFO] Compiler version: 3.3 144960ad
[INFO] Using provider: AXCLRTExecutionProvider
[INFO] SOC Name: AX650N
[INFO] VNPU type: VNPUType.DISABLED
[INFO] Compiler version: 3.3 0f7260e8
[INFO] Using provider: AXCLRTExecutionProvider
[INFO] SOC Name: AX650N
[INFO] VNPU type: VNPUType.DISABLED
[INFO] Compiler version: 3.3 144960ad
FaceAnalysisDIY warmup time: 0.213s
LandmarkRunner warmup time: 0.180s
2025-08-13 10:14:47.779 | INFO     | __main__:main:727 - Start making driving motion template...
2025-08-13 10:15:04.110 | INFO     | __main__:main:747 - Prepared pasteback mask done.
2025-08-13 10:15:04.903 | INFO     | __main__:main:785 - The animated video consists of 78 frames.
2025-08-13 10:15:11.468 | DEBUG    | __main__:warp_decode:647 - warp time: 6.561s
2025-08-13 10:25:50.630 | DEBUG    | __main__:warp_decode:647 - warp time: 8.343s
2025-08-13 10:25:51.493 | INFO     | __main__:has_audio_stream:114 - Error occurred while probing video: ./assets/examples/driving/d0.mp4, you may need to install ffprobe! (https://ffmpeg.org/download.html) Now set audio to false!
2025-08-13 10:25:54.608 | INFO     | __main__:main:870 - Animated video: ./axmodel_infer/s0--d0.mp4
2025-08-13 10:25:54.609 | INFO     | __main__:main:871 - Animated video with concat: ./axmodel_infer/s0--d0_concat.mp4
2025-08-13 10:25:54.644 | DEBUG    | __main__:<module>:894 - LivePortrait axmodel infer time: 670.930s 

元動画：

出力動画：

Next Overview

Linux PC

CM4Stack

CoreMP135

AI Accelerator Card

LLM-8850 Card

Quick Start

Vision Models

Large Language Models

Multimodal Models

Audio Models

Generative Models

Advanced Usage

LLM

リアルタイム音声アシスタント

OpenAI ボイスアシスタント

XiaoZhi ボイスアシスタント

XiaoLing ボイスアシスタント

AtomS3R-M12 Volcengine Kit

オフライン音声認識

Unit ASR

Home Assistant

Home Assistant OS

ESPHome

Industrial Control

StamPLC

IoT Measuring Instruments

Air Quality

Module13.2 PPS

VAMeter

T-Lite

Ezdata

Ethernet Camera

PoECAM

Wi-Fi Camera

TimerCAM

Unit CamS3/-5MP

AI Camera

UnitV2

M5StickV/UnitV

LoRa & LoRaWAN

TTN (The Things Network)

Meshtastic

Motor Control

Unit Roller485/CAN

Develop Tools

Hobby Kit

ファームウェアの初期化