Three-dimensional object detection from light detection and ranging (LiDAR) point clouds has gained great attention in recent years due to its wide applications in smart cities and autonomous driving. Cascade framework shows its advancement in 2-D object detection but is less investigated in 3-D space. Conventional cascade structures use multiple separate subnetworks to sequentially refine region proposals. Such methods, however, have limited ability to measure proposal quality in all the stages, and it is hard to achieve a desirable performance improvement in 3-D space. This article proposes a new cascade framework, termed Cascade Attention (CasA), for 3-D object detection from LiDAR point clouds. CasA consists of a region proposal network (RPN) and a cascade refinement network (CRN). In CRN, we designed a new cascade attention module (CAM) that uses multiple subnetworks and attention modules to aggregate the object features from different stages and progressively refine region proposals. CasA can be integrated into various two-stage 3-D detectors and improve their performance. Extensive experiments on the KITTI and Waymo datasets with various baseline detectors demonstrate the universality and superiority of our CasA. In particular, based on one variant of Voxel-region-based convolutional neural network (RCNN), we achieve the state-of-the-art results on the KITTI dataset. On the KITTI online 3-D object detection leaderboard, we achieve a high detection performance of 83.06%, 47.09%, and 73.47% average precision (AP) in the moderate car, pedestrian, and cyclist classes, respectively. Code is available at https://github.com/hailanyi/CasA

This paper proposes a new 3D multi-object tracker to more robustly track objects that are temporarily missed by detectors. Our tracker can better leverage object features for 3D Multi-Object Tracking (MOT) in point clouds. The proposed tracker is based on a novel data association scheme guided by prediction confidence, and it consists of two key parts. First, we design a new predictor that employs a constant acceleration (CA) motion model to estimate future positions, and outputs a prediction confidence to guide data association through increased awareness of detection quality. Second, we introduce a new aggregated pairwise cost to exploit features of objects in point clouds for faster and more accurate data association. The proposed cost consists of geometry, appearance and motion components. Specifically, we formulate the geometry cost using resolutions (lengths, widths and heights), centroids, and orientations of 3D bounding boxes (BBs), the appearance cost using appearance features from the deep learning-based detector backbone network, and the motion cost by associating different motion vectors. Extensive multi-object tracking experiments on the KITTI tracking benchmark demonstrated that our method outperforms, by a large margin, the state-of-the-art methods in both tracking accuracy and speed.

We present a novel deep convolutional network pipeline, LO-Net, for real-time lidar odometry estimation. Unlike most existing lidar odometry (LO) estimations that go through individually designed feature selection, feature matching, and pose estimation pipeline, LO-Net can be trained in an end-to-end manner. With a new mask-weighted geometric constraint loss, LO-Net can effectively learn feature representation for LO estimation, and can implicitly exploit the sequential dependencies and dynamics in the data. We also design a scan-to-map module, which uses the geometric and semantic information learned in LO-Net, to improve the estimation accuracy. Experiments on benchmark datasets demonstrate that LO-Net outperforms existing learning based approaches and has similar accuracy with the state-of-the-art geometry-based approach, LOAM.

This article presents a novel person reidentification model, named multihead self-attention network (MHSA-Net), to prune unimportant information and capture key local information from person images. MHSA-Net contains two main novel components: multihead self-attention branch (MHSAB) and attention competition mechanism (ACM). The MHSAB adaptively captures key local person information and then produces effective diversity embeddings of an image for the person matching. The ACM further helps filter out attention noise and nonkey information. Through extensive ablation studies, we verified that the MHSAB and ACM both contribute to the performance improvement of the MHSA-Net. Our MHSA-Net achieves competitive performance in the standard and occluded person Re-ID tasks.

Inpainting arbitrary missing regions is challenging because learning valid features for various masked regions is nontrivial. Though U-shaped encoder-decoder frameworks have been witnessed to be successful, most of them share a common drawback of mask unawareness in feature extraction because all convolution windows (or regions), including those with various shapes of missing pixels, are treated equally and filtered with fixed learned kernels. To this end, we propose our novel mask-aware inpainting solution. Firstly, a Mask-Aware Dynamic Filtering (MADF) module is designed to effectively learn multi-scale features for missing regions in the encoding phase. Specifically, filters for each convolution window are generated from features of the corresponding region of the mask. The second fold of mask awareness is achieved by adopting Point-wise Normalization (PN) in our decoding phase, considering that statistical natures of features at masked points differentiate from those of unmasked points. The proposed PN can tackle this issue by dynamically assigning point-wise scaling factor and bias. Lastly, our model is designed to be an end-to-end cascaded refinement one. Supervision information such as reconstruction loss, perceptual loss and total variation loss is incrementally leveraged to boost the inpainting results from coarse to fine. Effectiveness of the proposed framework is validated both quantitatively and qualitatively via extensive experiments on three public datasets including Places2, CelebA and Paris StreetView.