Fanyi Wang

Reconstruction of high-definition maps is a crucial task in perceiving the autonomous driving environment, as its accuracy directly impacts the reliability of prediction and planning capabilities in downstream modules. Current vectorized map reconstruction methods based on the DETR framework encounter limitations due to the redundancy in the decoder structure, necessitating the stacking of six decoder layers to maintain performance, which significantly hampers computational efficiency. To tackle this issue, we introduce FastMap, an innovative framework designed to reduce decoder redundancy in existing approaches. FastMap optimizes the decoder architecture by employing a single-layer, two-stage transformer that achieves multilevel representation capabilities. Our framework eliminates the conventional practice of randomly initializing queries and instead incorporates a heatmap-guided query generation module during the decoding phase, which effectively maps image features into structured query vectors using learnable positional encoding. Additionally, we propose a geometry-constrained point-to-line loss mechanism for FastMap, which adeptly addresses the challenge of distinguishing highly homogeneous features that often arise in traditional point-to-point loss computations. Extensive experiments demonstrate that FastMap achieves state-of-the-art performance in both nuScenes and Argoverse2 datasets, with its decoder operating 3.2 faster than the baseline. Code and more demos are available at https://github.com/hht1996ok/FastMap.

In the field of autonomous driving, online high-definition (HD) map reconstruction is crucial for planning tasks. Recent research has developed several high-performance HD map reconstruction models to meet this necessity. However, the point sequences within the instance vectors may be jittery or jagged due to prediction bias, which can impact subsequent tasks. Therefore, this paper proposes the Anti-disturbance Map reconstruction framework (ADMap). To mitigate point-order jitter, the framework consists of three modules: Multi-Scale Perception Neck, Instance Interactive Attention (IIA), and Vector Direction Difference Loss (VDDL). By exploring the point-order relationships between and within instances in a cascading manner, the model can monitor the point-order prediction process more effectively. ADMap achieves state-of-the-art performance on the nuScenes and Argoverse2 datasets. Extensive results demonstrate its ability to produce stable and reliable map elements in complex and changing driving scenarios. Code and more demos are available at https://github.com/hht1996ok/ADMap.

3D object detection is one of the most important tasks in autonomous driving and robotics. Our research focuses on tackling low efficiency issue of point-based methods on large-scale point clouds. Existing point-based methods adopt farthest point sampling (FPS) strategy for downsampling, which is computationally expensive in terms of inference time and memory consumption when the number of point cloud increases. In order to improve efficiency, we propose a novel Instance-Centroid Faster Point Sampling Module (IC-FPS) , which effectively replaces the first Set Abstraction (SA) layer that is extremely tedious. IC-FPS module is comprised of two methods, local feature diffusion based background point filter (LFDBF) and Centroid-Instance Sampling Strategy (CISS). LFDBF is constructed to exclude most invalid background points, while CISS substitutes FPS strategy by fast sampling centroids and instance points. IC-FPS module can be inserted to almost every point-based models. Extensive experiments on multiple public benchmarks have demonstrated the superiority of IC-FPS. On Waymo dataset, the proposed module significantly improves performance of baseline model and accelerates inference speed by 3.8 times. For the first time, real-time detection of point-based models in large-scale point cloud scenario is realized.

In the machine vision inspection of large-aperture aspheric optical components, the limited field of view and micron-level detection accuracy requirements make sub-aperture scanning imaging indispensable. High-precision scanning depends on the alignment of the spin axis of the mechanical system and the optical axis of the component, so the component needs to be centered before scanning. In view of this problem, this paper proposes a high-precision automatic centering method (HPACM) for rotationally symmetric aspheric optical elements based on machine vision. The goal is to adjust two reference points on the optical axis to the mechanical spin axis, which are the sphere center of the upper surface vertex and the sphere center of the lower surface vertex imaged by the upper surface of the element respectively. Adjust the first point to the spin axis is the Eccentric Error Correction (EEC), and then adjust the second point to the spin axis is the tilt error correction (TEC). In EEC and TEC, the component rotates one circle around the spin axis. If there exists eccentric or tilt error, the crosshair on the image plane will move along a circular trajectory which center on the spin axis. Then use pixel coordinates of crosshair center extraction algorithm (PCCEA) to extract the pixel coordinates of the centers of the crosshair images during the circular motion, and apply the least squares circle fitting algorithm to obtain the trajectory circle of the crosshair centers. The center of the crosshair and the center of the track circle on the image plane correspond to the sphere center of the surface vertex and a point on the spin axis of the object side respectively, and the relative positions of these two points on the image plane can be converted to the object side according to the system parameters. Experimental results show that the proposed HPACM can correct the eccentric and tilt error to within 7um and 0.5’.