An End-to-End System for Content-Based Video Retrieval Using Behavior, Actions, and Appearance with Interactive Query Refinement
| Autor: | Amit Srivastava, A. G. Amitha Perera, Linus Sherrill, Dong Liu, Kishore K. Reddy, Mita Desai, B.S. Manjunath, D. Rowley, Chia-Chih Chen, Anoop Kumar, Xiaoyang Wang, Bi Song, J. Liu, J. Luck, Chuck Atkins, Tsuhan Chen, D. Hanson, Roderic Collins, T. Rude, Amit K. Roy-Chowdhury, A. Jain, Daniel F. Keefe, Anthony Hoogs, Qiang Ji, Shih-Fu Chang, J. Kopaz, Rusty Blue, Arslan Basharat, Naresh P. Cuntoor, Keith Fieldhouse, Shiv Chandrasekaran, Eran Swears, C. Greco, Jelena Tesic, K. Chang, K. Sullivan, Zhaohui H. Sun, Larry S. Davis, Matthew Woehlke, B. Drew, Saurabh Khanwalkar, B. Boeckel, Mubarak Shah, Y. Yacoob, Jake K. Aggarwal |
|---|---|
| Rok vydání: | 2015 |
| Předmět: |
business.industry
Computer science ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Relevance feedback Basic Behavioral and Social Science Metadata Discriminative model Video tracking Behavioral and Social Science RGB color model Computer vision Artificial intelligence business Classifier (UML) Image resolution Graphical user interface |
| Zdroj: | AVSS |
| Popis: | We describe a system for content-based retrieval from large surveillance video archives, using behavior, action and appearance of objects. Objects are detected, tracked, and classified into broad categories. Their behavior and appearance are characterized by action detectors and descriptors, which are indexed in an archive. Queries can be posed as video exemplars, and the results can be refined through relevance feedback. The contributions of our system include the fusion of behavior and action detectors with appearance for matching; the improvement of query results through interactive query refinement (IQR), which learns a discriminative classifier online based on user feedback; and reasonable performance on low resolution, poor quality video. The system operates on video from ground cameras and aerial platforms, both RGB and IR. Performance is evaluated on publicly-available surveillance datasets, showing that subtle actions can be detected under difficult conditions, with reasonable improvement from IQR. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|