News
AI Summary
19 Nov 202610 Jumada II 1448 AH
Horizon Robotics Unveils LoFT Video Geometry Model

Horizon Robotics Unveils LoFT Video Geometry Model

Horizon Robotics has announced the LoFT model, which predicts camera pose, camera intrinsics, and per-frame depth from RGB video. This work was introduced on November 18 as a step towards enhancing spatial perception for robots. The LoFT model utilizes continuity across neighboring frames and information from different viewpoints to reconstruct a scene's visible three-dimensional structure. It features a two-loop Transformer architecture, where the first loop aggregates visual features across progressively wider local temporal windows, while the second loop updates the camera state.

Follow these topics

Sign in to follow the topics that matter to you

Sign in to follow

This summary is generated with AI and receives periodic editorial review. Refer to the original source for full details.

0
0 reading now

Insight Score

Rate to unlock

Sign in to react, rate, and save. Sign In

Key terms in this story

The top AI news daily on Telegram, and the week in your inbox.