Beyond Context: Exploring Semantic Similarity for Small Object Detection in Crowded Scenes


Beyond Context: Exploring Semantic Similarity for Small Object Detection in Crowded Scenes


報告題目:Beyond Context: Exploring Semantic Similarity for Small Object Detection in Crowded Scenes

報告人:Xiangjian He

單位:School of Electrical and Data Engineering, University of Technology Sydney




Professor Xiangjian He is the Director of Computer Vision and Pattern Recognition Laboratory at the Global Big Data Technologies Centre(GBDTC) at the University of Technology Sydney(UTS). He is an IEEE Senior Member and has been an IEEE Signal Processing Society Student Committee member. He received a UTS Chancellor’s Award for Research Excellence in 2018. He has also been awarded ‘Internationally Registered Technology Specialist’ by International Technology Institute(ITI). He has been carrying out research mainly in the areas of image processing, network security, pattern recognition, computer vision and machine learning in the previous years. He has played various chair roles in many international conferences such as ACM MM, MMM, ICDAR, IEEE BigDataSE, IEEE TrustCom, IEEE CIT, IEEE AVSS, IEEE TrustCom, IEEE ICPR and IEEE ICARCV. He has received many competitive national or regional grants awarded by Australian Research Council(ARC), National Natural Science Foundation of China(NSFC), Hong Kong Research Grants Council(RGC). Very recently, he has received an ARC-LP grant and industry grants awarded by Cisco, SAS, Sydney Trains, Data 61, RMCRC etc. In recent years, he has many high quality publications in prestigious journals such as Journal of the Association for Information Science and Technology and ACM Computing Surveys, IEEE Transactions journals such as IEEE Transactions on Dependable and Secure Computing, IEEE Transactions on Network Science and Engineering, IEEE Transactions on Mobile Computing, IEEE Transactions on Computers, IEEE Transactions on Parallel and Distributed Systems, IEEE Transactions on Multimedia, IEEE Transactions on Circuits and Systems for Video Technology, IEEE Transactions on Intelligent Transportation Systems, IEEE Transactions on Cloud Computing, IEEE Transactions on Reliability and IEEE Transactions on Consumer Electronics, and in Elsevier’s journals such as Pattern Recognition, Signal Processing, Automation in Construction, Information Sciences, Neurocomputing, Future Generation Computer Systems, Computer Networks, Computer and System Sciences, and Network and Computer Applications. He has also had papers published in premier international conferences and workshops such as ACL, IJCAI, CVPR, ECCV, ACM MM, TrustCom and WACV. 


Small object detection in crowded scene aims to find those tiny targets with very limited resolution from crowded scenes. Due to very little information available on tiny objects, it is often not suitable to detect them merely based on the information presented inside their bounding boxes, resulting low accuracy. In this talk, we exploit the semantic similarity among all predicted objects’ candidates to boost the performance of detectors when handling tiny objects. For this purpose, we construct a pairwise constraint to depict such semantic similarity and propose a new framework based on Discriminative Learning and Graph-Cut techniques. Experiments conducted on three widely used benchmark datasets demonstrate the improvement over the state-of-the-art approaches gained by applying this idea.