Generating city-scale lane-level maps faces significant challenges due to the intricate urban environments, such as blurred or absent lane markings. Additionally, a standard lane-level map requires a comprehensive organization of lane groupings, encompassing lane direction, style, boundary, and topology, yet has not been thoroughly examined in prior research. These obstacles result in labor-intensive human annotation and high maintenance costs. This paper overcomes these limitations and presents an industrial-grade solution named DuMapNet that outputs standardized, vectorized map elements and their topology in an end-to-end paradigm. To this end, we propose a group-wise lane prediction (GLP) system that outputs vectorized results of lane groups by meticulously tailoring a transformer-based network. Meanwhile, to enhance generalization in challenging scenarios, such as road wear and occlusions, as well as to improve global consistency, a contextual prompts encoder (CPE) module is proposed, which leverages the predicted results of spatial neighborhoods as contextual information. Extensive experiments conducted on large-scale real-world datasets demonstrate the superiority and effectiveness of DuMapNet. Additionally, DuMapNet has already been deployed in production at Baidu Maps since June 2023, supporting lane-level map generation tasks for over 360 cities while bringing a 95% reduction in costs. This demonstrates that DuMapNet serves as a practical and cost-effective industrial solution for city-scale lane-level map generation.

Offset-based representation has emerged as a promising approach for modeling semantic relations between pixels and object motion, demonstrating efficacy across various computer vision tasks. In this paper, we introduce a novel one-stage multi-tasking network tailored to extend the offset-based approach to MOTS. Our proposed framework, named OffsetNet, is designed to concurrently address amodal bounding box detection, instance segmentation, and tracking. It achieves this by formulating these three tasks within a unified pixel-offset-based representation, thereby achieving excellent efficiency and encouraging mutual collaborations. OffsetNet achieves several remarkable properties: first, the encoder is empowered by a novel Memory Enhanced Linear Self-Attention (MELSA) block to efficiently aggregate spatial-temporal features; second, all tasks are decoupled fairly using three lightweight decoders that operate in a one-shot manner; third, a novel cross-frame offsets prediction module is proposed to enhance the robustness of tracking against occlusions. With these merits, OffsetNet achieves 76.83% HOTA on KITTI MOTS benchmark, which is the best result without relying on 3D detection. Furthermore, OffsetNet achieves 74.83% HOTA at 50 FPS on the KITTI MOT benchmark, which is nearly 3.3 times faster than CenterTrack with better performance. We hope our approach will serve as a solid baseline and encourage future research in this field.

Recent advances in deep learning based algorithms have made it feasible to transfer image styles from an example image to other images. However, it is still hard to transfer the style of color sketches due to their unique texture statistics. In this paper, an automatic color sketch generation system is developed from existing real-time style transfer methods. We choose a suitable image from a set of carefully selected color sketch examples as the style target for every content image during training. We also propose a novel style transfer convolutional neural network with spatial refinement to realize high-resolution style transfer. Finally, gouache color is introduced to the generated images via a linear color transform followed by a guided filtering operation. Experimental results illustrate that our system can produce vivid color sketch images and greatly reduce artifacts compared to previous state-of-the-art methods.

Integrating LiDAR and camera information has been a widely adopted approach for 3D object detection in autonomous driving. Nevertheless, the unexplored potential of multi-modal fusion remains in the realm of offline 4D detection. We experimentally find that the root lies in two reasons: (1) the sparsity of point clouds poses a challenge in extracting long-term image features and thereby results in information loss. (2) some of the LiDAR points may be obstructed in the image, leading to incorrect image features. To tackle these problems, we first propose a simple yet effective offline multi-modal 3D object detection method, named Fusion4DAL, for 4D auto-labeling with long-term multi-modal sequences. Specifically, in order to address the sparsity of points within objects, we propose a multi-modal mixed feature fusion module (MMFF). In the MMFF module, we introduce virtual points based on a dense 3D grid and combine them with real LiDAR points. The mixed points are then utilized to extract dense point-level image features, thereby enhancing multi-modal feature fusion without being constrained by the sparse real LiDAR points. As to the obstructed LiDAR points, we leverage the occlusion relationship among objects to ensure depth consistency between LiDAR points and their corresponding depth feature maps, thus filtering out erroneous image features. In addition, we define a virtual point loss (VP Loss) to distinguish different types of mixed points and preserve the geometric shape of objects. Furthermore, in order to promote long-term receptive field and capture finer-grained features, we propose a global point attention decoder with a box-level self-attention module and a global point attention module. Finally, comprehensive experiments show that Fusion4DAL outperforms state-of-the-art offline 3D detection methods on nuScenes and Waymo dataset.