Time: 14.00-17.30, April 16th, 2019, Tuesday
Venue: Lecture Hall, 9th floor, College of Data Science
Report 1
Title: A Preliminary Study Deep Cognition Neural Networks
Speaker: Liang Wang
A researcher and doctoral supervisor at Institute of Automation, Chinese Academy of Sciences
Member of International Association of Pattern Recognition (IAPR Fellow, 2014)
Member of Institute of Electrical and Electronics Engineers (IEEE Fellow, 2019)
Abstract:
Aiming at the frontier of pattern recognition and cognitive science, and breaking the limitations of existing neural networks in terms of structure and function, we hope to create new models and methods of deep cognitive neural network that integrate feedback, attention, memory and other visual cognition inspired mechanisms and apply them to the task of visual pattern analysis. This report will focus on the recent exploratory work on deep cognitive neural networks.
Report 2
Title: Intelligent Analysis and Information Mining of Remote Sensing Image
Speaker: Guisong Xia
Professor and doctoral supervisor at Wuhan University
IEEE senior member
Abstract:
The observation network covering space, sky and land has accumulated rich remote sensing image data. With the improvement of image resolution, remote sensing images contain increasing information. One of the research frontiers in current remote sensing information analysis is how to effectively analyze remote sensing image and how to find key ground features and geological knowledge from the images. Facing the bottleneck in applying remote sensing images and combining the current emerging technologies of computer vision and machine learning, this report will analyze such issues as feature extraction, scene classification, semantic segmentation, and target detection and recognition. Mr. Xia’s research on these issues include: the image database construction method for large-scale remote sensing image information mining, the efficient feature computing model for high-resolution remote sensing images, the machine learning method for scene classification of large-scale remote sensing images and target detection and recognition.
Report 3
Title: Intelligent Image Computing
Speaker: Jiaying Liu, associate professor at Peking University
Abstract:
There is a rapid development in Image automatic stylization respect to intelligent video editing and generation. This report will introduce the technologies of character stylization based on statistical features, unsupervised stylization of characters and automatic generation of pictures and texts, textual stylization and de-stylization based on generative adversarial networks.
Report 4
Title: Visual Object Tracking Via Deep Regression Model
Speaker: Chao Ma, assistant professor, Shanghai Jiao Tong University
Abstract:
Visual object tracking is challenging as target objects often undergo significant appearance changes. In this talk, I will present our work on how to best exploit deep regression networks to improve tracking accuracy and robustness. First, I will introduce our TPAMI 2018 (ICCV 2015) work, in which we adaptively learn correlation filters on hierarchical convolutional layers to precisely locate targets. Second, I will present our ICCV 2017 work, where we reformulate correlation filters by a one-layer neural network. We additionally exploit the spatial and temporal residual learning scheme to facilitate visual tracking. Last, I will report our recent work in ECCV 2018. In this work, we propose a novel shrinkage loss to train deep regression networks for visual tracking. Extensive experimental results on large scale benchmark datasets show that the proposed algorithms perform favorably against state-of-the-art methods.
Translated and Edited by Weiwei Wang
Go Back