论文标题
4D无监督的对象发现
4D Unsupervised Object Discovery
论文作者
论文摘要
对象发现是计算机视觉中的核心任务。尽管在监督的对象检测中已经取得了快速进展，但其无监督的对应物在很大程度上尚未探索。随着数据量的增长，注释的昂贵成本是阻碍进一步研究的主要限制。因此，发现没有注释的物体具有重要意义。但是，由于缺乏歧视性信息，这项任务似乎在静止图像或点云上似乎是不切实际的。先前的研究忽略了多模式输入背后的关键时间信息和约束。在本文中，我们提出了4D无监督的对象发现，并从4D数据中共同发现对象-3D点云和2D RGB图像具有时间信息。我们通过在3D点云上提出一个clusternet，介绍了该任务的第一种实际方法，该clusternet通过2D本地化网络共同迭代优化。大规模Waymo打开数据集的广泛实验表明，本地化网络和clusternet在类别不稳定的2D对象检测和3D实例分段上都能达到竞争性能，从而弥合了无概括方法和完全监督的方法之间的差距。代码和模型将在https://github.com/robertwyq/lsmol上提供。
Object discovery is a core task in computer vision. While fast progresses have been made in supervised object detection, its unsupervised counterpart remains largely unexplored. With the growth of data volume, the expensive cost of annotations is the major limitation hindering further study. Therefore, discovering objects without annotations has great significance. However, this task seems impractical on still-image or point cloud alone due to the lack of discriminative information. Previous studies underlook the crucial temporal information and constraints naturally behind multi-modal inputs. In this paper, we propose 4D unsupervised object discovery, jointly discovering objects from 4D data -- 3D point clouds and 2D RGB images with temporal information. We present the first practical approach for this task by proposing a ClusterNet on 3D point clouds, which is jointly iteratively optimized with a 2D localization network. Extensive experiments on the large-scale Waymo Open Dataset suggest that the localization network and ClusterNet achieve competitive performance on both class-agnostic 2D object detection and 3D instance segmentation, bridging the gap between unsupervised methods and full supervised ones. Codes and models will be made available at https://github.com/Robertwyq/LSMOL.
