PulseAugur
EN
LIVE 09:11:17
中文(ZH) 登顶 ICML Oral !专访上交大团队:这个 3D 自动标注 AI 太强了

AI system Holi-Spatial generates 3D annotations from 2D video without human input

A team of Chinese researchers has developed Holi-Spatial, an AI system that automatically generates 3D spatial data annotations from ordinary 2D videos without human intervention. This system bypasses the need for expensive hardware like LiDAR scanners and the labor-intensive process of manual 3D bounding box annotation. Holi-Spatial utilizes a three-stage pipeline involving geometric optimization, image-level perception with large language models, and scene-level refinement to create structured 3D data, addressing a decade-long bottleneck in the field. AI

IMPACT Automates 3D data annotation, potentially accelerating research and development in spatial intelligence and embodied AI.

RANK_REASON The item describes a new AI system and its technical pipeline for generating 3D spatial data, which has been accepted as an oral presentation at a major AI conference (ICML). [lever_c_demoted from research: ic=1 ai=1.0]

Read on 雷峰网 (Leiphone) →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

AI system Holi-Spatial generates 3D annotations from 2D video without human input

COVERAGE [1]

  1. 雷峰网 (Leiphone) TIER_1 中文(ZH) ·

    Topping ICML Oral! Exclusive Interview with SJTU Team: This 3D Auto-Labeling AI is Too Powerful

    <section style="text-align: center; margin: 0px 16px; line-height: 1.75em; display: block;"><img class="rich_pages wxw-img" src="https://static.leiphone.com/uploads/new/images/20260715/6a56fd758bf5c.jpg?imageMogr2/quality/90" style="width: 100%; display: inline-block; text-align:…