Timezone: »

Test-Time Personalization with a Transformer for Human Pose Estimation
Yizhuo Li · Miao Hao · Zonglin Di · Nitesh Bharadwaj Gundavarapu · Xiaolong Wang

Tue Dec 07 08:30 AM -- 10:00 AM (PST) @

We propose to personalize a 2D human pose estimator given a set of test images of a person without using any manual annotations. While there is a significant advancement in human pose estimation, it is still very challenging for a model to generalize to different unknown environments and unseen persons. Instead of using a fixed model for every test case, we adapt our pose estimator during test time to exploit person-specific information. We first train our model on diverse data with both a supervised and a self-supervised pose estimation objectives jointly. We use a Transformer model to build a transformation between the self-supervised keypoints and the supervised keypoints. During test time, we personalize and adapt our model by fine-tuning with the self-supervised objective. The pose is then improved by transforming the updated self-supervised keypoints. We experiment with multiple datasets and show significant improvements on pose estimations with our self-supervised personalization. Project page with code is available at https://liyz15.github.io/TTP/.

Author Information

Yizhuo Li (Shanghai Jiao Tong University)
Miao Hao (University of California, San Diego)
Zonglin Di (University of California, San Diego)
Nitesh Bharadwaj Gundavarapu
Xiaolong Wang (UC San Diego)

More from the Same Authors