In this paper, we propose a novel system named Disp R-CNN for 3D object detection from stereo images. Many recent works solve this problem by first recovering point clouds with disparity estimation and then apply a 3D detector. The disparity map is computed for the entire image, which is costly and fails to leverage category-specific prior. In contrast, we design an instance disparity estimation network (iDispNet) that predicts disparity only for pixels on objects of interest and learns a category-specific shape prior for more accurate disparity estimation. To address the challenge from scarcity of disparity annotation in training, we propose to use a statistical shape model to generate dense disparity pseudo-ground-truth without the need of LiDAR point clouds, which makes our system more widely applicable. Experiments on the KITTI dataset show that, when LiDAR ground-truth is not used at training time, Disp R-CNN outperforms previous state-of-the-art methods based on stereo input by 20% in terms of average precision for all categories. The code and pseudo-ground-truth data are available at the project page: https://github.com/zju3dv/disprcnn.

中文翻译：

---

用于3D对象检测的Shape Prior指导实例视差估计。

在本文中，我们提出了一种名为Disp R-CNN的新颖系统，用于从立体图像中检测3D对象。许多最近的工作通过首先使用视差估计恢复点云，然后应用3D检测器来解决此问题。对于整个图像计算视差图，这是昂贵的并且不能利用特定于类别的先验。相反，我们设计了一个实例视差估计网络（iDispNet），该网络仅预测感兴趣对象上像素的视差，并在获得特定类别的形状之前先进行更精确的视差估计。为了解决培训中视差标注的稀缺性带来的挑战，我们建议使用统计形状模型来生成密集的视差伪地面真相，而无需使用LiDAR点云，这使得我们的系统更广泛地适用。在KITTI数据集上进行的实验表明，在训练时不使用LiDAR地面真相时，Disp R-CNN在所有类别的平均精度方面都比基于立体声输入的最新技术要高20％。代码和伪地面真相数据可在项目页面上找到：https：//github.com/zju3dv/disprcnn。