三维物体感知 Transformer（3PT）
3D-Object Perception Transformer (3PT)
该工作把检测、分割与 6DoF 姿态估计统一进一个无需深度输入、依靠多视图信息持续精化物体表征的 Transformer 框架，可直接作为零样本 3D 物体感知的强基线，替代冻结基础模型集成式流水线，对机器人抓取与工业装配等下游视觉任务有直接复用价值。
原文依据 3PT continuously refines its object representations without depth input, enhancing 3D understanding by incorporating multi-view information.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论当前的零样本三维物体感知方法通常依赖多个冻结基础模型的集成。这限制了对物体的深层理解与跨域泛化能力，使其性能不足以支撑真实世界部署。三维物体感知 Transformer（3PT）通过将检测、分割与 6DoF 姿态估计统一到单一框架中，并直接针对三维物体感知进行训练，来解决这一局限。3PT 基于两个分别专注于二维与三维以物体为中心场景理解的大规模训练 Transformer，在无深度输入的情况下持续精化其物体表征，并通过融合多视图信息增强三维理解。3PT 在 BOP 基准上的检测与姿态估计达到最优水平，常常取得两位数的提升，在许多情况下优于非零样本方法，并在 BOP-2025 挑战赛的 11 个赛道中赢得 7 个。3PT 在检测与姿态估计上超越任务专用模型，在多样的 BOP 基准上常常取得两位数的百分比提升，在某些情况下优于非零样本方法。它还在 BOP Challenge 2025 的 11 个赛道中排名第一的有 7 个。3PT 的高精度与可靠性非常适合诸如料箱抓取与精密插入等实际工业机器人应用。项目主页见 https://www.intrinsic.ai/publications/3pt-cvpr2026。
英文原文摘要 1433 字符
Current approaches to zero-shot 3D-object perception typically rely on ensembles of frozen foundation models. This limits deep object understanding and cross-domain generalization, making performance inadequate for real-world deployment. The 3D-Object Perception Transformer (3PT) addresses this limitation by unifying detection, segmentation, and 6DoF pose estimation in a single framework, directly trained for 3D-object perception. Based on two large-scale trained transformers that specialize in 2D and 3D object-centric scene understanding respectively, 3PT continuously refines its object representations without depth input, enhancing 3D understanding by incorporating multi-view information. 3PT is the state-of-the-art for detection and pose estimation on the BOP benchmarks, often achieving double digit improvements, in many cases, outperforming non-zero-shot methods, and winning 7 of 11 tracks in the BOP-2025 challenge. 3PT surpasses task-specialized models for detection and pose estimation, often achieving double-digit percentage improvements on the diverse BOP-benchmarks, and in some cases outperforming non zero-shot methods. It also ranked first in 7 of 11 tracks at the BOP Challenge 2025. 3PT's high-accuracy and reliability is well-suited for practical industrial robotics applications such as bin picking and precise insertion. Project Page can be found at https://www.intrinsic.ai/publications/3pt-cvpr2026.