arXiv subjects

Xiaokai Bai

Publications and source records attributed to Xiaokai Bai.

22 records · Page 2Linked to original sources

arXiv2025

Boosting Multi-View Indoor 3D Object Detection via Adaptive 3D Volume Construction

This work presents SGCDet, a novel multi-view indoor 3D object detection framework based on adaptive 3D volume construction. Unlike previous approaches that restrict the receptive field of voxels to fixed locations on images, we introduce a geometry and context aware aggregation module to integrate geometric and contextual information within adaptive regions in each image and dynamically adjust the contributions from different views, enhancing the representation capability of voxel features. Furthermore, we propose a sparse volume construction strategy that adaptively identifies and selects voxels with high occupancy probabilities for feature refinement, minimizing redundant computation in free space. Benefiting from the above designs, our framework achieves effective and efficient volume construction in an adaptive way. Better still, our network can be supervised using only 3D bounding boxes, eliminating the dependence on ground-truth scene geometry. Experimental results demonstrate that SGCDet achieves state-of-the-art performance on the ScanNet, ScanNet200 and ARKitScenes datasets. The source code is available at https://github.com/RM-Zhang/SGCDet.

cs.CV↗

arXiv2025

Super high capacity of silicon carbon anode over 6500 mAh g-1 for lithium battery

As silicon is approaching its theoretical limit for the anode materials in lithium battery, searching for a higher limit is indispensable. Herein, we demonstrate the possible of achieving ultrahigh capacity over 6500 mAh g-1 in silicon-carbon composites. Considering the numerous defects inside the silicon nanostructures, it is deduced the formation of quasi-Bose Einstein condensation should be possible, which can lead to the low viscosity flow of lithium-ions through the anode. At a charge-discharge rate of 0.1C (0.42 A g-1), the initial discharge specific capacity reaches 6694.21 mAh g-1, with a Coulomb efficiency (CE) of 74.71%, significantly exceeding the theoretical capacity limit of silicon. Further optimization of the anode material ratio results in improved cycling stability, with a discharge specific capacity of 5542.98 mAh g-1 and a CE of 85.25% at 0.1C. When the initial discharge capacity is 4043.01 mAh g-1, the CE rises to 86.13%. By training a multilayer perceptron with material parameters as inputs and subsequently optimizing it using a constrained genetic algorithm, an initial discharge specific capacity of up to 7789.55 mAh g-1 can be achieved theoretically. This study demonstrates that silicon-carbon composites have great potential to significantly enhance the energy density of lithium-ion batteries.

physics.chem-ph↗

arXiv2025

Structure-Aware Radar-Camera Depth Estimation

Radar has gained much attention in autonomous driving due to its accessibility and robustness. However, its standalone application for depth perception is constrained by issues of sparsity and noise. Radar-camera depth estimation offers a more promising complementary solution. Despite significant progress, current approaches fail to produce satisfactory dense depth maps, due to the unsatisfactory processing of the sparse and noisy radar data. They constrain the regions of interest for radar points in rigid rectangular regions, which may introduce unexpected errors and confusions. To address these issues, we develop a structure-aware strategy for radar depth enhancement, which provides more targeted regions of interest by leveraging the structural priors of RGB images. Furthermore, we design a Multi-Scale Structure Guided Network to enhance radar features and preserve detailed structures, achieving accurate and structure-detailed dense metric depth estimation. Building on these, we propose a structure-aware radar-camera depth estimation framework, named SA-RCD. Extensive experiments demonstrate that our SA-RCD achieves state-of-the-art performance on the nuScenes dataset. Our code will be available at https://github.com/FreyZhangYeh/SA-RCD.

cs.CV↗

arXiv2024

Rethinking Early-Fusion Strategies for Improved Multispectral Object Detection

Most recent multispectral object detectors employ a two-branch structure to extract features from RGB and thermal images. While the two-branch structure achieves better performance than a single-branch structure, it overlooks inference efficiency. This conflict is increasingly aggressive, as recent works solely pursue higher performance rather than both performance and efficiency. In this paper, we address this issue by improving the performance of efficient single-branch structures. We revisit the reasons causing the performance gap between these structures. For the first time, we reveal the information interference problem in the naive early-fusion strategy adopted by previous single-branch structures. Besides, we find that the domain gap between multispectral images, and weak feature representation of the single-branch structure are also key obstacles for performance. Focusing on these three problems, we propose corresponding solutions, including a novel shape-priority early-fusion strategy, a weakly supervised learning method, and a core knowledge distillation technique. Experiments demonstrate that single-branch networks equipped with these three contributions achieve significant performance enhancements while retaining high efficiency. Our code will be available at \url{https://github.com/XueZ-phd/Efficient-RGB-T-Early-Fusion-Detection}.

cs.CV↗