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

Despite the promising performance of current 3D human pose estimationtechniques, understanding and enhancing their generalization on challengingin-the-wild videos remain an open problem. In this work, we focus on therobustness of 2D-to-3D pose lifters. To this end, we develop two benchmarkdatasets, namely Human3.6M-C and HumanEva-I-C, to examine the robustness ofvideo-based 3D pose lifters to a wide range of common video corruptionsincluding temporary occlusion, motion blur, and pixel-level noise. We observethe poor generalization of state-of-the-art 3D pose lifters in the presence ofcorruption and establish two techniques to tackle this issue. First, weintroduce Temporal Additive Gaussian Noise (TAGN) as a simple yet effective 2Dinput pose data augmentation. Additionally, to incorporate the confidencescores output by the 2D pose detectors, we design a confidence-awareconvolution (CA-Conv) block. Extensively tested on corrupted videos, theproposed strategies consistently boost the robustness of 3D pose lifters andserve as new baselines for future research.

Improving the Robustness of 3D Human Pose Estimation: A Benchmark and Learning from Noisy Input