RTMPose-S (ONNX)

RTMPose-S pose-estimation model converted to ONNX, for use with OpenCV dnn / FaceDetectorYN-style pipelines and the react-native-opencv demo.

Source: OpenMMLab RTMPose-S (via the PINTO0309 model-zoo ONNX conversion).
License: Apache-2.0 (RTMPose is © OpenMMLab, Apache-2.0; this mirror is redistribution with attribution).
Input: 192x256 (W×H), float32, mean [123.675, 116.28, 103.53], std ~`57.5. Output: SimCC simcc_x(384) /simcc_y` (512), COCO-17 kpts.
File: rtmpose-s_working.onnx, 21,889,761 bytes, sha256 d869413ccfbc0ed027337c773f124d6362436239a83fd0a6e2d43b7a8377792e.

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support