Spaces:

alexyywwdd

/

postometro-free-demo

like 11

Sleeping

We want to apply for a community grant to demonstrate our academic project "PostoMETRO: Pose Token Enhanced Mesh Transformer for Robust 3D Human Mesh Recovery".
With the recent advancements in single-image-based human mesh recovery, there is a growing interest in enhancing its performance in certain extreme scenarios, such as occlusion, while maintaining overall model accuracy.
Our proposed PostoMETRO (POSe TOken enhanced MEsh TRansfOrmer), which integrates occlusion-resilient 2D pose representation into transformers in a token-wise manner, achieves SOTA performance on both standard and occlusion-specific benchmarks.
Our method requires 18GB VRAM, hence we apply for an A10G GPU for running our demo.

Arxiv link: https://arxiv.org/abs/2403.12473

Some qualitative results: