Attention-DP3：通过几何对齐的注意力条件实现空间物体感知的3D扩散策略
Attention-DP3: Spatially Object-aware 3D Diffusion Policy via Geometry-aligned Attentional Conditioning
浏览论文内容
中文总结 AI 辅助
提出Attention-DP3，通过三场注意力条件化注入物体级几何线索，提升3D扩散策略在杂乱场景中的目标定位能力，在多个基准上达到最先进性能，尤其在高干扰下比DP3提升高达31%。
中文摘要 AI 辅助
在复杂、杂乱的操作场景中，3D点云观测本质上具有模糊性，目标物体可能被部分遮挡或与视觉上相似的干扰物紧密混杂。因此，随着场景复杂度的增加，标准的3D扩散策略往往难以定位和利用任务相关的几何信息。我们提出了Attention-DP3，一种空间物体感知的3D扩散策略，通过注意力注入物体级几何线索，同时保持DP3扩散骨干网络不变。我们的流程在RGB图像上执行开放词汇的2D分割，然后利用校准的相机几何将预测的目标掩码提升到3D，以获得以物体为中心的几何先验。我们通过三场注意力条件化（Tri-field Attentional Conditioning）整合这些线索，该机制构建三个互补场：（i）目标性场（targetness field）用于锚定目标物体，（ii）目标内显著性场（intra-target saliency field）用于强调目标内与任务相关的几何，（iii）背景性场（backgroundness field）用于抑制干扰物和杂乱。在Adroit、DexArt、MetaWorld以及真实世界的SO101平台上的实验表明，与DP3相比，我们取得了一致的改进，在多个基准上达到了最先进的性能。值得注意的是，随着干扰物数量的增加，DP3的性能急剧下降，而Attention-DP3保持稳定，在重度杂乱环境下比DP3高出最多31%。代码已在此https URL公开。
英文摘要
3D point-cloud observations are inherently ambiguous in complex, cluttered manipulation scenes, where target objects may be partially occluded or tightly intermingled with visually similar distractors. As a result, standard 3D diffusion policies often struggle to localize and exploit task-relevant geometry as scene complexity grows. We propose \textbf{Attention-DP3}, a spatially object-aware 3D diffusion policy that injects object-level geometric cues via attention while keeping the DP3 diffusion backbone unchanged. Our pipeline performs open-vocabulary 2D segmentation on RGB images, then lifts predicted target masks into 3D using calibrated camera geometry to obtain object-centric geometric priors. We incorporate these cues through Tri-field Attentional Conditioning, which constructs three complementary fields: (i) a targetness field to anchor the target object, (ii) an intra-target saliency field to emphasize task-relevant geometry within the target, and (iii) a backgroundness field to suppress distractors and clutter. Experiments on Adroit, DexArt, MetaWorld, and the real-world SO101 platform show consistent improvements over DP3, achieving state-of-the-art performance across benchmarks. Notably, as distractor objects increase, DP3 drops sharply, whereas Attention-DP3 remains stable and outperforms DP3 by up to 31\% under heavy clutter. The code is publicly available at https://github.com/zhangzhongbo2213/Attention-DP3.
发表机构
- Dalian University of Technology（大连理工大学）
机构由 AI 辅助整理，请以论文原文为准。
