Detection of Key Points in Mice at Different Scales via Convolutional Neural Network

In this work, we propose a symmetry approach and design a convolutional neural network for mouse pose estimation under scale variation. The backbone adopts the UNet structure, uses the residual network to extract features, and adds the ASPP module into the appropriate residual units to expand the perceptual field, and uses the deep and shallow feature fusion to fuse and process the features at multiple scales to capture the various spatial relationships related to body parts to improve the recognition accuracy of the model. Finally, a set of prediction results based on heat map and coordinate offset is generated. We used our own built mouse dataset and obtained state-of-the-art results on the dataset.

Automated summary

In this study, we propose a symmetry approach and design a convolutional neural network for mouse pose estimation under scale variation. The network employs the UNet structure as the backbone, utilizes residual network for feature extraction, incorporates the ASPP module to expand the perceptual field, and integrates deep and shallow features to capture spatial relationships at multiple scales to optimize recognition accuracy. Experimental results show that our method achieves state-of-the-art performance on our self-built mouse dataset.