English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In this work, we propose a new solution to 3D human pose estimation invideos. Instead of directly regressing the 3D joint locations, we drawinspiration from the human skeleton anatomy and decompose the task into bonedirection prediction and bone length prediction, from which the 3D jointlocations can be completely derived. Our motivation is the fact that the bonelengths of a human skeleton remain consistent across time. This promotes us todevelop effective techniques to utilize global information across all theframes in a video for high-accuracy bone length prediction. Moreover, for thebone direction prediction network, we propose a fully-convolutional propagatingarchitecture with long skip connections. Essentially, it predicts thedirections of different bones hierarchically without using any time-consumingmemory units e.g. LSTM). A novel joint shift loss is further introduced tobridge the training of the bone length and bone direction prediction networks.Finally, we employ an implicit attention mechanism to feed the 2D keypointvisibility scores into the model as extra guidance, which significantlymitigates the depth ambiguity in many challenging poses. Our full modeloutperforms the previous best results on Human3.6M and MPI-INF-3DHP datasets,where comprehensive evaluation validates the effectiveness of our model.

Anatomy-aware 3D Human Pose Estimation with Bone-based Pose Decomposition