Edge devices, with their widely varying capabilities, support a diverse range of edge AI models. This raises the question: how does an edge model differ from a high-accuracy (base) model for the same task? We introduce XDELTA, a novel explainable AI tool that explains differences between a high-accuracy base model and a computationally efficient but lower-accuracy edge model. To achieve this, we propose a learning-based approach to characterize the model difference, named the DELTA network, which complements the feature representation capability of the edge network in a compact form. To construct DELTA, we propose a sparsity optimization framework that extracts the essence of the base model to ensure compactness and sufficient feature representation capability of DELTA, and implement a negative correlation learning approach to ensure it complements the edge model. We conduct a comprehensive evaluation to test XDELTA's ability to explain model discrepancies, using over 1.2 million images and 24 models, and assessing real-world deployments with six participants. XDELTA excels in explaining differences between base and edge models (arbitrary pairs as well as compressed base models) through geometric and concept-level analysis, proving effective in real-world applications.
Rethinking Disparity: A Depth Range Free Multi-View Stereo Based on Disparity
Wuhan University(武汉大学);The Hong Kong University of Science and Technology(香港科技大学);Harbin Institute of Technology (Shenzhen)(哈尔滨工业大学(深圳));XGRIDS;The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州))
Multiple Lane Detection Algorithm Based on Optimised Dense Disparity Map Estimation
Tsinghua University(清华大学);Shanghai Jiao Tong University(上海交通大学);The Hong Kong University of Science and Technology(香港科技大学);University of Bristol(布里斯托尔大学);Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences(中国科学院深圳先进技术研究院)
Pothole Detection Based on Disparity Transformation and Road Surface Modeling
Hong Kong University of Science and Technology(香港科技大学);Near East University(近东大学);National Oceanography Centre(国家海洋学中心);Aristotle University of Thessaloniki(塞萨洛尼基亚里士多德大学)
DispBench: Benchmarking Disparity Estimation to Synthetic Corruptions
Data and Web Science Group, University of Mannheim(曼海姆大学数据与网络科学小组);Max-Planck-Institute for Informatics, Saarland Informatics Campus(马克斯·普朗克研究所信息学研究所萨尔兰信息学校区)
本文提出DispBench基准测试工具,系统评估视差估计方法在对抗攻击和2D Common Corruptions等合成损坏下的鲁棒性,并通过迄今最广泛的分析揭示准确率、可靠性与泛化能力之间的关键相关性。
Bidirectional skip-frame prediction for video anomaly detection with intra-domain disparity-driven attention
School of Computer Science and Engineering, Xi'an University of Technology, Xi'an, China(西安理工大学计算机科学与工程学院);CCTEG Changzhou Research Institute(中技集团常州研究院;天迪(常州)自动化有限公司);Tiandi(Changzhou) Automation Co., Ltd.(西安理工大学计算机科学与技术学院);College of Computer Science and Technology, Xi'an University of Science and Technology
Graph Attention Layer Evolves Semantic Segmentation for Road Pothole Detection: A Benchmark and Algorithms
Tongji University(同济大学);Shanghai Research Institute for Intelligent Autonomous Systems(上海智能自主系统研究院);the Hong Kong University of Science and Technology(香港科技大学);SmartMore(思谋科技);University of Thessaloniki(塞萨洛尼基大学)
Reducing Class-Wise Performance Disparity via Margin Regularization
Nanyang Technological University(南洋理工大学);Hefei University of Technology(合肥工业大学);Singapore Management University(新加坡管理学院);University of Science and Technology of China(中国科学技术大学)