Publications (25)
Improving Neural Indoor Surface Reconstruction with Mask-Guided Adaptive Consistency Constraints
Xinyi Yu, Liqin Lu, Jintao Rong +2
3D scene reconstruction from 2D images has been a long-standing task. Instead of estimating per-frame depth maps and fusing them in 3D, recent research leverages the neural implici…
When Distillation Breaks Motion Control: Restoring Generative Trajectories for Fast Video Generators
Jintao Rong, Xin Xie, Xinyi Yu +4
Training-free motion customization imposes motion patterns from reference videos onto video generators through test-time computation. Most existing methods target full diffusion mo…
Retrieval-Enhanced Visual Prompt Learning for Few-shot Classification
Jintao Rong, Hao Chen, Linlin Ou +3
The Contrastive Language-Image Pretraining (CLIP) model has been widely used in various downstream vision tasks. The few-shot learning paradigm has been widely adopted to augment i…
ShiftNAS: Improving One-shot NAS via Probability Shift
Mingyang Zhang, Xinyi Yu, Haodong Zhao +1
One-shot Neural architecture search (One-shot NAS) has been proposed as a time-efficient approach to obtain optimal subnet architectures and weights under different complexity case…
Explicit Solution of Tunable Input-to-State Safe-Based Controller Under High-Relative-Degree Constraints
Yan Wei, Yu Feng, Linlin Ou +2
This paper investigates the safety analysis and verification of nonlinear systems subject to high-relative-degree constraints and unknown disturbance. The closed-form solution of t…
Efficient Re-parameterization Operations Search for Easy-to-Deploy Network Based on Directional Evolutionary Strategy
Xinyi Yu, Xiaowei Wang, Jintao Rong +2
Structural re-parameterization (Rep) methods has achieved significant performance improvement on traditional convolutional network. Most current Rep methods rely on prior knowledge…
MKIoU Loss: Towards Accurate Oriented Object Detection in Aerial Images
Xinyi Yu, Jiangping Lu, Mi Lin +1
Oriented bounding box regression is crucial for oriented object detection. However, regression-based methods often suffer from boundary problems and the inconsistency between loss…
Pedestrian Attribute Recognition in Video Surveillance Scenarios Based on View-attribute Attention Localization
Weichen Chen, Xinyi Yu, Linlin Ou
Pedestrian attribute recognition in surveillance scenarios is still a challenging task due to the inaccurate localization of specific attributes. In this paper, we propose a novel…
Real-time Rail Recognition Based on 3D Point Clouds
Xinyi Yu, Weiqi He, Xuecheng Qian +2
Accurate rail location is a crucial part in the railway support driving system for safety monitoring. LiDAR can obtain point clouds that carry 3D information for the railway enviro…
A Self-adaptive SAC-PID Control Approach based on Reinforcement Learning for Mobile Robots
Xinyi Yu, Yuehai Fan, Siyu Xu +1
Proportional-integral-derivative (PID) control is the most widely used in industrial control, robot control and other fields. However, traditional PID control is not competent when…
Multi-subgoal Robot Navigation in Crowds with History Information and Interactions
Xinyi Yu, Jianan Hu, Yuehai Fan +2
Robot navigation in dynamic environments shared with humans is an important but challenging task, which suffers from performance deterioration as the crowd grows. In this paper, mu…
Across-Task Neural Architecture Search via Meta Learning
Jingtao Rong, Xinyi Yu, Mingyang Zhang +1
Adequate labeled data and expensive compute resources are the prerequisites for the success of neural architecture search(NAS). It is challenging to apply NAS in meta-learning scen…
3DGSNav: Enhancing Vision-Language Model Reasoning for Object Navigation via Active 3D Gaussian Splatting
Wancai Zheng, Hao Chen, Xianlong Lu +2
Object navigation is a core capability of embodied intelligence, enabling an agent to locate target objects in unknown environments. Recent advances in vision-language models (VLMs…
CrossFusion: Interleaving Cross-modal Complementation for Noise-resistant 3D Object Detection
Yang Yang, Weijie Ma, Hao Chen +2
The combination of LiDAR and camera modalities is proven to be necessary and typical for 3D object detection according to recent studies. Existing fusion strategies tend to overly…
Conditional Generative Data-free Knowledge Distillation
Xinyi Yu, Ling Yan, Yang Yang +2
Knowledge distillation has made remarkable achievements in model compression. However, most existing methods require the original training data, which is usually unavailable due to…
RepNAS: Searching for Efficient Re-parameterizing Blocks
Mingyang Zhang, Xinyi Yu, Jingtao Rong +1
In the past years, significant improvements in the field of neural architecture search(NAS) have been made. However, it is still challenging to search for efficient networks due to…
Effective Model Compression via Stage-wise Pruning
Mingyang Zhang, Xinyi Yu, Jingtao Rong +1
Automated Machine Learning(Auto-ML) pruning methods aim at searching a pruning strategy automatically to reduce the computational complexity of deep Convolutional Neural Networks(d…
GSORB-SLAM: Gaussian Splatting SLAM benefits from ORB features and Transmittance information
Wancai Zheng, Xinyi Yu, Jintao Rong +3
The emergence of 3D Gaussian Splatting (3DGS) has recently ignited a renewed wave of research in dense visual SLAM. However, existing approaches encounter challenges, including sen…
Graph Pruning for Model Compression
Mingyang Zhang, Xinyi Yu, Jingtao Rong +1
Previous AutoML pruning works utilized individual layer features to automatically prune filters. We analyze the correlation for two layers from the different blocks which have a sh…
UP-SLAM: Adaptively Structured Gaussian SLAM with Uncertainty Prediction in Dynamic Environments
Wancai Zheng, Linlin Ou, Jiajie He +3
Recent 3D Gaussian Splatting (3DGS) techniques for Visual Simultaneous Localization and Mapping (SLAM) have significantly progressed in tracking and high-fidelity mapping. However,…
LoRAPrune: Structured Pruning Meets Low-Rank Parameter-Efficient Fine-Tuning
Mingyang Zhang, Hao Chen, Chunhua Shen +4
Large Language Models (LLMs), such as LLaMA and T5, have shown exceptional performance across various tasks through fine-tuning. Although low-rank adaption (LoRA) has emerged to ch…
Channel Merging: Preserving Specialization for Merged Experts
Mingyang Zhang, Jing Liu, Ganggui Ding +3
Lately, the practice of utilizing task-specific fine-tuning has been implemented to improve the performance of large language models (LLM) in subsequent tasks. Through the integrat…
Boosting Box-supervised Instance Segmentation with Pseudo Depth
Xinyi Yu, Ling Yan, Pengtao Jiang +4
The realm of Weakly Supervised Instance Segmentation (WSIS) under box supervision has garnered substantial attention, showcasing remarkable advancements in recent years. However, t…
Oriented Object Detection in Aerial Images Based on Area Ratio of Parallelogram
Xinyi Yu, Mi Lin, Jiangping Lu +1
Oriented object detection is a challenging task in aerial images since the objects in aerial images are displayed in arbitrary directions and are frequently densely packed. The mai…
A Self-adaptive LSAC-PID Approach based on Lyapunov Reward Shaping for Mobile Robots
Xinyi Yu, Siyu Xu, Yuehai Fan +1
To solve the coupling problem of control loops and the adaptive parameter tuning problem in the multi-input multi-output (MIMO) PID control system, a self-adaptive LSAC-PID algorit…