Jie Qin

Disco4D: Towards Prior-Free Motion Representation Learning for Point Cloud Videos Via Self-Disentangled Contrastive Pre-Training

Zhi Zuo, Chenyi Zhuang, Pan Gao, Jie Qin et autres

cn, it (code pays fourni par la source)

Predicting fuzzy topological indices from crisp indices in hexagonal and honeycomb networks using linear regression

Jie Qin, Shamaila Yousaf, Mamoona Waris, Anisa Naeem et autres

Classical (crisp) graph models fail to capture uncertainty inherent in real-world networks, whereas fuzzy topological indices provide a richer description but are computationally expensive. This study investigates whether fuzzy topological indices can be accurately predicted from their crisp counterparts using machine learning. …

cn, pk (code pays fourni par la source)

Dual-view Driver Gaze Estimation via Mutual Enhancement

Chengzheng Fu, Rong Quan, Siyu Chen, Yiming Ni et autres

cn (code pays fourni par la source)

PRISM: Preference-Guided Semantic Reasoning with Vision-Language Models for Object Goal Navigation

Zijian Liu, Cong Pan, Chengjie Fan, Wanjie Cai et autres

Object Goal Navigation (ObjectNav) requires an agent to locate a target object in unseen environments based on partial visual observations. A central challenge lies in reasoning about where the target is likely to appear under semantic uncertainty, sparse observations, and ambiguous scene …

cn (code pays fourni par la source)

Object Referring-Guided Scanpath Prediction with Perception-Enhanced Vision-Language Models

Rong Quan, Yantao Lai, Dong Liang, Jie Qin

cn (code pays fourni par la source)

Object Referring-Guided Scanpath Prediction with Perception-Enhanced Vision-Language Models

Rong Quan, Yantao Lai, Dong Liang, Jie Qin

Object Referring-guided Scanpath Prediction (ORSP) aims to predict the human attention scanpath when they search for a specific target object in a visual scene according to a linguistic description describing the object. Multimodal information fusion is a key point of ORSP. Therefore, …

cn (code pays fourni par la source)

DP-Retinex: Dual-Prior Guided Low-Light Image Enhancement With YUV-Domain Reflectance-Illumination Decomposition

Zhengkai Zhao, Longmi Gao, Pan Gao, Jie Qin

Accurate estimation of reflectance and illumination maps in the Retinex framework remains a significant challenge for low-light image enhancement due to inherent decomposition ambiguity. To address this, we propose DualPrior-Retinex, a novel framework that, inspired by Retinex theory, leverages the practical advantages …

cn (code pays fourni par la source)

ACGFormer: Attribute Classification Guided Transformer for Camouflaged Object Detection

Wutao Liu, Yao Yuan, Pan Gao, Zheng Lin et autres

cn (code pays fourni par la source)

History-Enhanced Two-Stage Transformer for Aerial Vision-and-Language Navigation

Xichen Ding, Jianzhe Gao, Cong Pan, Wenguan Wang et autres

Aerial Vision-and-Language Navigation (AVLN) requires Unmanned Aerial Vehicle (UAV) agents to localize targets in large-scale urban environments based on linguistic instructions. While successful navigation demands both global environmental reasoning and local scene comprehension, existing UAV agents typically adopt mono-granularity frameworks that struggle …

STAR: STacked AutoRegressive Scheme for Unified Multimodal Learning

Jie Qin, Jiancheng Huang, Limeng Qiao, Lin Ma

Multimodal large language models (MLLMs) play a pivotal role in advancing the quest for general artificial intelligence. However, achieving unified target for multimodal understanding and generation remains challenging due to optimization conflicts and performance trade-offs. To effectively enhance generative performance while preserving …

DCN: Decoupled-Coupled Network for Text-based Person Search

Zhiyuan Qi, Rong Quan, L.i-Ming Su, Yizhen Jia et autres

Text-based person search aims to identify a person based on textual descriptions, by simultaneously addressing person detection and cross-modal alignment between text queries and person images. Existing approaches often struggle with conflicts in exploiting proposals across these two sub-tasks. Specifically, cross-modal alignment …

cn (code pays fourni par la source)

A versatile strategy of wire electric discharge machining for one-step fabrication of molybdenum-based micro-supercapacitors: from MoOx to CuMoOx

Ri Chen, Jie Qin, Ziliang Chen, Jinyang Liu et autres

cn, ca (code pays fourni par la source)