https://www.solosoft.dev/tags/ant-group/Recent content in Ant Group on SoloSoftHugoen-ushttps://www.solosoft.dev/post/lingbot-map-3d-reconstruction-2026/Mon, 01 Jan 0001 00:00:00 +0000https://www.solosoft.dev/post/lingbot-map-3d-reconstruction-2026/<p>3D scene reconstruction has long been a foundational challenge in computer vision. Traditional approaches rely on expensive LiDAR hardware, offline batch processing, or iterative optimization that is too slow for real-time applications. On April 16, 2026, <strong>Robbyant</strong> &ndash; the embodied AI division of Ant Group (蚂蚁集团) &ndash; released <strong>LingBot-Map</strong> (<a href="https://github.com/robbyant/lingbot-map">github.com/robbyant/lingbot-map</a>), a feed-forward 3D foundation model that changes this equation entirely.</p> <p>LingBot-Map takes a single RGB video stream and reconstructs dense, accurate 3D environments in real time &ndash; no LiDAR, no multi-pass optimization, no offline processing. It runs at approximately 20 FPS at 518x378 resolution and maintains consistent accuracy over sequences exceeding 10,000 frames. The paper, available on arXiv (<a href="https://arxiv.org/abs/2604.14141">2604.14141</a>), reports state-of-the-art results across multiple benchmarks, including an Absolute Trajectory Error (ATE) of 6.42 meters on the Oxford Spires dataset &ndash; a 2.8x improvement over prior methods &ndash; and an F1 score of 98.98 on ETH3D, more than 20 points ahead of the competition.</p>