Search

Online high-definition (HD) map construction is an important and challenging task in autonomous driving. Recently, there has been a growing interest in cost-effective multi-view camera-based methods without relying on other sensors like LiDAR. However, these methods suffer from a lack of explicit depth information, necessitating the use of large models to achieve satisfactory performance. To address this, we employ the Knowledge Distillation (KD) idea for efficient HD map construction for the first time and introduce a novel KD-based approach called MapDistill to transfer knowledge from a high-performance camera-LiDAR fusion model to a lightweight camera-only model. Specifically, we adopt the teacher-student architecture, i.e., a camera-LiDAR fusion model as the teacher and a lightweight camera model as the student, and devise a dual BEV transform module to facilitate cross-modal knowledge distillation while maintaining cost-effective camera-only deployment. Additionally, we present a comprehensive distillation scheme encompassing cross-modal relation distillation, dual-level feature distillation, and map head distillation. This approach alleviates knowledge transfer challenges between modalities, enabling the student model to learn improved feature representations for HD map construction. Experimental results on the challenging nuScenes dataset demonstrate the effectiveness of MapDistill, surpassing existing competitors by over 7.7 mAP or 4.5X speedup., Comment: Accepted by ECCV2024

Various models based on StyleGAN have gained significant traction in the field of image synthesis, attributed to their robust training stability and superior performances. Within the StyleGAN framework, the adoption of image skip connection is favored over the traditional residual connection. However, this preference is just based on empirical observations; there has not been any in-depth mathematical analysis on it yet. To rectify this situation, this brief aims to elucidate the mathematical meaning of the image skip connection and introduce a groundbreaking methodology, termed the image squeeze connection, which significantly improves the quality of image synthesis. Specifically, we analyze the image skip connection technique to reveal its problem and introduce the proposed method which not only effectively boosts the GAN performance but also reduces the required number of network parameters. Extensive experiments on various datasets demonstrate that the proposed method consistently enhances the performance of state-of-the-art models based on StyleGAN. We believe that our findings represent a vital advancement in the field of image synthesis, suggesting a novel direction for future research and applications.

We present a global optimization routine for the variational quantum algorithms, which utilizes the dynamic tunneling flow. Originally designed to leverage information gathered by a gradient-based optimizer around local minima, we adapt the conventional dynamic tunneling flow to exploit the distance measure of quantum states, resolving issues of extrinsic degeneracy arising from the parametrization of quantum states. Our global optimization algorithm is applied to the variational quantum eigensolver for the transverse-field Ising model to demonstrate the performance of our routine while comparing it with the conventional dynamic tunneling method, which is based on the Euclidean distance measure on the parameter space., Comment: 12 pages; 6 figures

Vectorized High-Definition (HD) map construction requires predictions of the category and point coordinates of map elements (e.g. road boundary, lane divider, pedestrian crossing, etc.). State-of-the-art methods are mainly based on point-level representation learning for regressing accurate point coordinates. However, this pipeline has limitations in obtaining element-level information and handling element-level failures, e.g. erroneous element shape or entanglement between elements. To tackle the above issues, we propose a simple yet effective HybrId framework named HIMap to sufficiently learn and interact both point-level and element-level information. Concretely, we introduce a hybrid representation called HIQuery to represent all map elements, and propose a point-element interactor to interactively extract and encode the hybrid information of elements, e.g. point position and element shape, into the HIQuery. Additionally, we present a point-element consistency constraint to enhance the consistency between the point-level and element-level information. Finally, the output point-element integrated HIQuery can be directly converted into map elements' class, point coordinates, and mask. We conduct extensive experiments and consistently outperform previous methods on both nuScenes and Argoverse2 datasets. Notably, our method achieves $77.8$ mAP on the nuScenes dataset, remarkably superior to previous SOTAs by $8.3$ mAP at least., Comment: Accepted to CVPR 2024

In reinforcement learning, temporal abstraction in the action space, exemplified by action repetition, is a technique to facilitate policy learning through extended actions. However, a primary limitation in previous studies of action repetition is its potential to degrade performance, particularly when sub-optimal actions are repeated. This issue often negates the advantages of action repetition. To address this, we propose a novel algorithm named Uncertainty-aware Temporal Extension (UTE). UTE employs ensemble methods to accurately measure uncertainty during action extension. This feature allows policies to strategically choose between emphasizing exploration or adopting an uncertainty-averse approach, tailored to their specific needs. We demonstrate the effectiveness of UTE through experiments in Gridworld and Atari 2600 environments. Our findings show that UTE outperforms existing action repetition algorithms, effectively mitigating their inherent limitations and significantly enhancing policy learning efficiency., Comment: Accepted in AAAI 2024 (Main Technical Track)

The interplay between ferromagnetism and the non-trivial topology has unveiled intriguing phases in the transport of charges and spins. For example, it is consistently observed the so-called topological Hall effect (THE) featuring a hump structure in the curve of the Hall resistance (Rxy) vs. a magnetic field (H) of a heterostructure consisting of a ferromagnet (FM) and a topological insulator (TI). The origin of the hump structure is still controversial between the topological Hall effect model and the multi-component anomalous Hall effect (AHE) model. In this work, we have investigated a heterostructure consisting of BixSb2-xTeySe3-y (BSTS) and Cr2Te3 (CT), which are well-known TI and two-dimensional FM, respectively. By using the so-called minor-loop measurement, we have found that the hump structure observed in the CT/BSTS is more likely to originate from two AHE channels. Moreover, by analyzing the scaling behavior of each amplitude of two AHE with the longitudinal resistivities of CT and BSTS, we have found that one AHE is attributed to the extrinsic contribution of CT while the other is due to the intrinsic contribution of BSTS. It implies that the proximity-induced ferromagnetic layer inside BSTS serves as a source of the intrinsic AHE, resulting in the hump structure explained by the two AHE model.

We present a systematic Raman spectroscopy study on a series of pyrochlore ruthenates, a system which is not yet clearly settled on its magnetic origin and structure. Apart from the Raman-active phonon modes, new peaks that appear in the energy range of 15 - 35 meV below the N\'{e}el temperature are assigned as one-magnon modes. The temperature evolution of one-magnon modes displays no significant thermal dependence in mode frequencies while the intensities decrease monotonically. Remarkably, one-magnons from all compounds show similar characteristics with a single dominant peak at lower energy and weaker side peaks at a couple of meV higher energy. Most importantly, we uncover a striking proportionality between the dominant magnon mode energies and the N\'{e}el temperature values. Our results suggest the Ru ions may have similar or the same magnetic phase in all pyrochlore ruthenates of our study. We have thus found an avenue for directly tuning the magnetic exchange interaction by the selection of the $A$-site ion.