A Large-Scale Mouse Pose Dataset for Mouse Pose Estimation

Mouse pose estimations have important applications in the fields of animal behavior research, biomedicine, and conservation studies. Accurate efficient mouse using computer vision are necessary. Although methods for developed, bottlenecks still exist. One most prominent problems is lack uniform standardized training datasets. Here, we resolve this difficulty by introducing dataset. Our dataset contains 40,000 frames RGB images large-scale 2D ground-truth motion images. All were captured from interacting lab mice through a stable single viewpoint, including 5 distinct species 20 total. Moreover, to improve annotation efficiency, five keypoints creatively proposed, which one keypoint at center other two pairs symmetric. Then, created simple, yet effective software that works annotating It another link establish benchmark model estimations. We employed modified object detections estimation algorithms achieve precise, effective, robust performances. As first large dataset, our proposed will help advance research on assist application areas related experiments.