# HKU open-sources GenSfM: jointly reconstructing 3D objects and camera poses from four uncalibrated images

- Source: 3D视觉工坊 · 微信公众号
- Published: 2026-10-30T23:00:00.000Z
- RoboHot: https://robohot.news/en/items/zrozjuvv8u6ircukenm85sd3z
- Original: https://mp.weixin.qq.com/s?__biz=MzU1MjY4MTA1MQ==&mid=2247752710&idx=1&sn=5766336915505db45f705941f3bac14a&chksm=fa4decdfa1904ee5b98fc1a2fd54c5c78272194014170a55d79314cec424cabc27d0e44b5d95&scene=126&sessionid=1793442626#rd

## Summary

A University of Hong Kong team has open-sourced GenSfM, which uses a native 3D diffusion model to jointly denoise object shape and per-view camera poses within a single multimodal sequence, requiring only 2 to 4 uncalibrated images and no camera intrinsics or extrinsics. Evaluated on Toys4k and GSO, it improves novel-view PSNR by 3.2–4.5 dB over the multi-view baseline TRELLIS-M, reduces median Chamfer distance by about 64%–74%, and achieves pose Acc@5° above 96%.

## Why it matters

The work turns a native 3D diffusion model into the main inference mechanism for joint denoising of shape and camera poses, and releases code and models as a reference for sparse-view reconstruction.