René's URL Explorer Experiment

Title: Fix RTMPose likelihood computation by maximpavliv · Pull Request #3078 · DeepLabCut/DeepLabCut · GitHub

Open Graph Title: Fix RTMPose likelihood computation by maximpavliv · Pull Request #3078 · DeepLabCut/DeepLabCut

X Title: Fix RTMPose likelihood computation by maximpavliv · Pull Request #3078 · DeepLabCut/DeepLabCut

Description: Background The likelihoods (confidences) produced by SimCCPredictor were not consistent with the expected behavior of SimCC decoding. In some experiments, likelihoods exceeded 1, because the predictor was returning raw model outputs (logits) instead of proper probabilities. When setting apply_softmax=True, likelihoods were extremely small (~1e-3). This is expected mathematically (softmax distributes probability mass across the entire SimCC space), but not useful as a visibility/confidence measure. When setting normalize_outputs=True, most likelihoods were exactly 1, because normalization simply divided by the maximum logit value. This mismatch originated from how the raw SimCC representation vectors (logits over discretized x/y coordinates) were converted into coordinates and visibility scores. See the original MMPose predictor implementation for reference. Changes This PR brings SimCCPredictor closer to the original MMPose decoding logic: 1. Added parameters sigma: variance of the Gaussian used to generate SimCC labels during training. This controls the expected sharpness of the label distribution. decode_beta: scaling factor applied together with sigma to logits before softmax. It acts like an inverse temperature in softmax: sharpening peaked distributions (visible keypoints) and flattening diffuse ones (occluded keypoints). Together, sigma * decode_beta ensures that visibility/confidence is decoded consistently with the label distribution used in training. 2. Likelihood computation If normalize_outputs=False (which is the default): the raw logits are scaled by (sigma * decode_beta) before calling get_simcc_maximum with apply_softmax=True. This matches the MMPose decoding step and yields meaningful visibility/confidence scores in [0,1]. 3. Default behavior apply_softmax now defaults to True (to ensure proper probabilities rather than raw logits are returned). normalize_outputs still defaults to False. Outcome Likelihoods are now properly bounded in [0,1]. The values reflect both the sharpness of the SimCC distribution and the visibility of the keypoint, consistent with the SimCC paper and MMPose implementation. Users can still enable normalize_outputs for compatibility with previous behavior, but it is discouraged for real confidence estimation.

Open Graph Description: Background The likelihoods (confidences) produced by SimCCPredictor were not consistent with the expected behavior of SimCC decoding. In some experiments, likelihoods exceeded 1, because the predi...

X Description: Background The likelihoods (confidences) produced by SimCCPredictor were not consistent with the expected behavior of SimCC decoding. In some experiments, likelihoods exceeded 1, because the predi...

Opengraph URL: https://github.com/DeepLabCut/DeepLabCut/pull/3078

X: @github

direct link

Domain: github.com

Links:

Viewport: width=device-width