！！！有问题，欢迎扫码关注公众号/加群【392784757】一起交流讨论！！！

环境说明【开箱即用】

算力平台：https://passport.compshare.cn/register?referral_code=3u8UGQXls2tB7CoxKBKdEi ubuntu 22.04 python 3.12 pytorch 2.9.1+cu130

内置基础模型(0.4b seg/pose 0.1b pretrain) 位于 /workspace/sapiens2_host

各模型独立下载，使用软链接 链接到/workspace/sapiens2_host 下即可；/workspace/sapiens2_host 被软链接到 {HOME}/sapiens2_host

check 组织形式 应如下

sapiens2_host/
├── pretrain/
│   ├── sapiens2_{0.1b,0.4b,0.8b,1b,5b}_pretrain.safetensors
│   └── sapiens2_1b_4k_pretrain.safetensors
├── pose/
│   └── sapiens2_{0.4b,0.8b,1b,5b}_pose.safetensors
├── seg/
│   └── sapiens2_{0.4b,0.8b,1b,5b}_seg.safetensors
├── normal/
│   └── sapiens2_{0.4b,0.8b,1b,5b}_normal.safetensors
├── pointmap/
│   └── sapiens2_{0.4b,0.8b,1b,5b}_pointmap.safetensors
├── matting/
│   └── sapiens2_1b_matting.safetensors
└── detector/                  # [optional] only needed for pose inference
    └── detr-resnet-101-dc5/

快速验证

export SAPIENS_ROOT=/workspace/sapiens2
cd $SAPIENS_ROOT/sapiens/dense
./scripts/demo/seg.sh

Distributing 100 image paths into 1 jobs.
Model loaded from /root/sapiens2_host/seg/sapiens2_0.4b_seg.safetensors
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 100/100 [00:26<00:00,  3.73it/s]
Output directory: /workspace/sapiens2/seg-output/sapiens2_0.4b

0.4b 显存占用 大概在 3.2G

Wed Jul 22 15:48:10 2026
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 595.80                 Driver Version: 595.80         CUDA Version: 13.2     |
+-----------------------------------------+------------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA GeForce RTX 4090        Off |   00000000:C1:00.0 Off |                  Off |
| 47%   44C    P2            265W /  450W |    3204MiB /  24564MiB |    100%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI              PID   Type   Process name                        GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|    0   N/A  N/A             581      C   python                                 3194MiB |
+-----------------------------------------------------------------------------------------+