Categorical Depth Distribution Network for Monocular 3D Object Detection

Abstract

Monocular 3D object detection is a key problem for autonomous vehicles, as it provides a solution with simple configuration compared to typical multi-sensor systems. The main challenge in monocular 3D detection lies in accurately predicting object depth, which must be inferred from object and scene cues due to the lack of direct range measurement. Many methods attempt to directly estimate depth to assist in 3D detection, but show limited performance as a result of depth inaccuracy. Our proposed solution, Categorical Depth Distribution Network (CaDDN), uses a predicted categorical depth distribution for each pixel to project rich contextual feature information to the appropriate depth interval in 3D space. We then use the computationally efficient bird's-eye-view projection and single-stage detector to produce the final output detections. We design CaDDN as a fully differentiable end-to-end approach for joint depth estimation and object detection. We validate our approach on the KITTI 3D object detection benchmark, where we rank 1 st among published monocular methods. We also provide the first monocular 3D detection results on the newly released Waymo Open Dataset. We provide a code release for CaDDN which is made available here.

Geometry Uncertainty Projection Network for Monocular 3D Object Detection

“…Furthermore, our method also outperforms the methods with extra data. For instance, compared with the recently proposed CaDNN [38] utilizing LiDAR signals as supervision of depth estimation sub-task, our method still obtains 0.94%, 0.79%, and 0.31% gains on the three difficulty settings, which confirms the effectiveness of the proposed method.…”

Multimodal Virtual Point 3D Detection

“…A quantitative comparison between 2D and 3D detectors is included in Section 5.2. methods. On state-of-the-art 3D detection benchmarks [2,12], state-of-the-art monocular methods [26,41] achieve about half the mAP detection accuracy, of standard Lidar based baselines [60]. Pseudo-Lidar [56] based methods produce a virtual point cloud from RGB images, similar to our approach.…”

Densely Constrained Depth Estimator for Monocular 3D Object Detection

“…We also achieve state-of-the-art performance on WOD [38] as Table 3 shows. We surpass the previous state-of-the-art methods such as CaDDN [31] and PCT [41]. We also re-implement MonoFlex [51] on WOD [38] for a fair comparison.…”

“…Italics: These methods utilize the whole train set, while the others use 1/3 amount of images in train set. ‡: M3D-RPN is reimplemented by [31]. †: PatchNet is re-implemented by [41].…”