DWPOSE_MODEL_NAME if pose_estimator == "dw-ll_ucoco_384.onnx" else "hr16/UnJIT-DWPose", pose_repo, yolo_repo, cache_dir=annotator_ckpts_path, det_filename=bbox_detector, pose_filename=pose_estimator ) detect_hand = detect_hand == "enable" detect_body = detect_body == "enable"...
目前看主要原因是缺了如下两个文件: dw-ll_ucoco_384.onnxyolox_l.onnx 这两个文件的手动下载地址: https://huggingface.co/yzd-v/DWPose/tree/main 下完之后拷贝到如下位置: stable-diffusion-webui/extensions/sd-webui-controlnet/annotator/downloads/openpose 然后就看起来OK了,虽然我觉得这个手有7根手...
path.join(model_base_path, "dw-ll_ucoco_384.onnx"), device=device) self.dwprocessor = DWposeDetector( model_det=self.det, model_pose=self.pose)ref_image = ref_image.squeeze(0).cpu().numpy() * 255self.det = self.det.to(device)...
First, you need to download our Pose model dw-ll_ucoco_384.onnx (baidu,google) and Det model yolox_l.onnx (baidu,google), then put them into ControlNet-v1-1-nightly/annotator/ckpts. Then you can use DWPose to generate the images you like. ...
I also found that, there are noDWPosefolder in extensions. But, there was an empty anDWPosefolder in the directory"stable-diffusion-webui\tmp". Somehow,"dw-ll_ucoco_384.onnx"and"yolox_l.onnx"are present in"stable-diffusion-webui\extensions\sd-webui-controlnet\annotator\downloads\openpos...
Microsoft.ML.OnnxRuntime (1.15.1) 【原项目用1.15.1转的模型,不清楚降低版本后还能否使用】 OpencvSharp4 (4.8) 【版本实际上无限制】 模型文件 从https://huggingface.co/yzd-v/DWPose下载: dw-ll_ucoco_384.onnx yolox_l.onnx 使用样例
./pretrained_weights/ |-- DWPose | |-- dw-ll_ucoco_384.onnx | `-- yolox_l.onnx |-- image_encoder | |-- config.json | `-- pytorch_model.bin |-- denoising_unet.pth |-- motion_module.pth |-- pose_guider.pth |-- reference_unet.pth |-- sd-vae-ft-mse | |-- config.js...
First, you need to download our Pose model dw-ll_ucoco_384.onnx (baidu,google) and Det model yolox_l.onnx (baidu,google), then put them into ControlNet-v1-1-nightly/annotator/ckpts. Then you can use DWPose to generate the images you like. ...
def estimate_pose(self, image, detect_hand="enable", detect_body="enable", detect_face="enable", resolution=512, bbox_detector="yolox_l.onnx", pose_estimator="dw-ll_ucoco_384.onnx", **kwargs): def estimate_pose(self, image, detect_hand="enable", detect_body="enable", detect...
目前看主要原因是缺了如下两个文件: dw-ll_ucoco_384.onnxyolox_l.onnx 这两个文件的手动下载地址: https://huggingface.co/yzd-v/DWPose/tree/main 下完之后拷贝到如下位置: stable-diffusion-webui/extensions/sd-webui-controlnet/annotator/downloads/openpose 然后就看起来OK了,虽然我觉得这个手有7根手...