OpenXRLab

A one-stop open source platform for solving algorithm needs in the XR field

[OpenXRLab @OpenXRLab]

🥳XRNeRF v0.7.0 has been released!
- Add a new scene-nerf method: BungeeNerf.
- Support a new data format: Multiscale Google.
- Support highly modular NeRF methods implementation.

👉⚗Welcome to try at https://t.co/2NrQmXexjA

[OpenXRLab @OpenXRLab]

#HuMMan dataset v1.0 has released Reconstruction Subset!

XRMoCap supports SMPL estimation toolchain of #HuMMan dataset.

[Ziwei Liu @liuziwei7]

We released #HuMMan dataset v1.0 with Reconstruction Subset

- Subjects with diverse appearances and actions

- Multi-view images, masks (via matting), SMPL parameters and camera parameters are provided

- Benchmark NeRF algorithms under realistic settings with commercial sensors

[OpenXRLab @OpenXRLab]

🤩XRMoCap v0.7.0 has been released!
- Add multi-view multi-person inference.
- Add multiple optimizers, grad clipping, detailed joint weights and priors for better SMPLify results.
- Support using mediapipe for 2d perception.

👉⚗️Welcome to try at https://t.co/kC47dh8qnK

[OpenXRLab @OpenXRLab]

We welcome any issues and PRs! Feel free to try it out on @github : https://t.co/041toRLBGc

[SenseTime @SenseTime_AI]

#OpenXRLab, the open-source platform for #XR newly launched 1 basic and 6 code repositories, including 3 in spatial computing, 2 in human-computer interaction and 1 in rendering generation.

Follow @OpenXRLab on @Github: https://t.co/EhI8QCqkVO