Simple Baselines for Human Pose Estimation and Tracking

Abstract

There has been significant progress on pose estimation and increasing interests on pose tracking in recent years. At the same time, the overall algorithm and system complexity increases as well, making the algorithm analysis and comparison more difficult. This work provides simple and effective baseline methods. They are helpful for inspiring and evaluating new ideas for the field. State-of-the-art results are achieved on challenging benchmarks. The code will be available at https://github. com/leoxiaobin/pose.pytorch.

Cross View Fusion for 3D Human Pose Estimation

“…We adopt the network proposed in [31] as our base network and use ResNet-152 as its backbone, which was pretrained on the ImageNet classification dataset. The input Table 1.…”

“…Table 1 shows the results on the most important joints when we train, either only on the H36M dataset, or on a combination of the H36M and MPII datasets. It compares our approach with the baseline method [31], termed Single, which does not perform cross view feature fusion. We also compare with two baselines which compute sum or max values over the epipolar line using the camera parameters.…”

“…We conduct experiments on the H36M dataset using NO pose annotations. The single view pose estimator [31] is trained on the MPII dataset. If we directly apply this model to the test set of H36M and estimate the 3D pose by RPSM, the MPJPE error is about 109mm.…”

FollowMeUp Sports: New Benchmark for 2D Human Keypoint Recognition

“…[22] utilises a Symmetric Spatial Transformer Network to handle inaccurate bounding boxes. [24] uses simple deconvolution layers to obtain high-resolution heatmaps for human pose estimation. On the side of bottom-up methods, [9] proposes a limb descriptor and an efficient bottom-up grouping approach to associate neighbouring joints.…”

Multi-Level Network for High-Speed Multi-Person Pose Estimation

“…A lot of practical problems, such as smart environments, human-computer interaction [1], augmented reality [2], virtual reality, human behaviour analysis and recognition [3], require the information of human body keypoints. Similar to many vision problems, deep learning based methods in the problem of human pose estimation had significant progress in the past few years [4]. While the problem of human pose estimation is in quick developing, we also notice that the network architecture has a tendency of becoming more deep and complex.…”