Live viewers
A live look at who's streaming these scenes. Each arc links our lab at ETH Zürich to a viewer, fading as it settles — including yours, right now. (Pages are actually served worldwide via CDN; the Zürich node is our symbolic home.)
● you ● origin · ETH Zürich
Establishing connection…
Citation
📄 SmoothMotionVectors: Optimizing Your Content for Video Codecs in Free View Video Compression
SIGGRAPH Conference Papers ’26, Los Angeles, CA, USA · DOI 10.1145/3799902.3811112
ACM Reference Format
Mingyang Song, Yang Zhang, Siyu Tang, and Tunç Ozan Aydın. 2026. SmoothMotionVectors: Optimizing Your Content for Video Codecs in Free View Video Compression. In Special Interest Group on Computer Graphics and Interactive Techniques Conference Conference Papers (SIGGRAPH Conference Papers ’26), July 19–23, 2026, Los Angeles, CA, USA. ACM, New York, NY, USA, 11 pages. https://doi.org/10.1145/3799902.3811112
BibTeX
@inproceedings{song2026smoothmotionvectors,
author = {Song, Mingyang and Zhang, Yang and Tang, Siyu and Ayd{\i}n, Tun\c{c} Ozan},
title = {SmoothMotionVectors: Optimizing Your Content for Video Codecs in Free View Video Compression},
year = {2026},
isbn = {979-8-4007-2554-8/26/07},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3799902.3811112},
doi = {10.1145/3799902.3811112},
booktitle = {Special Interest Group on Computer Graphics and Interactive Techniques Conference Conference Papers (SIGGRAPH Conference Papers '26)},
numpages = {11},
location = {Los Angeles, CA, USA},
series = {SIGGRAPH Conference Papers '26}
}
Built with
Our contribution is the compressor — the in-browser player itself is only possible thanks to antimatter15's splat WebGL2 rasterizer (after EWA splatting & 3D Gaussian Splatting), FFmpeg via ffmpeg.wasm for in-browser HEVC decode, and fflate for keyframe NPZ unzip.
Datasets
The sample scenes come from open-source dynamic-scene datasets, used here for research demonstration. Please cite the original authors:
D-NeRF — synthetic dynamic scenes.
Albert Pumarola, Enric Corona, Gerard Pons-Moll, and Francesc Moreno-Noguer. 2021. D-NeRF: Neural Radiance Fields for Dynamic Scenes. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).
HyperNeRF — casually captured real-world scenes.
Keunhong Park, Utkarsh Sinha, Peter Hedman, Jonathan T. Barron, Sofien Bouaziz, Dan B Goldman, Ricardo Martin-Brualla, and Steven M. Seitz. 2021. HyperNeRF: A Higher-Dimensional Representation for Topologically Varying Neural Radiance Fields. ACM Transactions on Graphics (Proc. SIGGRAPH Asia) 40, 6 (2021).
NeRF-DS — real-world captures of dynamic specular objects.
Zhiwen Yan, Chen Li, and Gim Hee Lee. 2023. NeRF-DS: Neural Radiance Fields for Dynamic Specular Objects. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).
🔄 An evolving gallery — we'll keep adding scenes to these datasets, with more datasets to come. Check back for new results.
Note: this lightweight in-browser pipeline may differ slightly from the original GPU rendering — small differences arise from the CPU depth sort and the simplified rasterizer.
Please contact Mingyang Song (mingyang.song@inf.ethz.ch) for questions.