IEEE First Workshop on Emergent Issues in Large Amounts of Visual Data (WS-LAVD) October 4, 2009 in Kyoto, Japan. 9:30-9:40 Opening Remarks Session: Segmentation and Clustering 9:40-10:55, 25x3 min 9:40-10:05 Applying Incremental Learning to Parallel Image Segmentation, Cyril Charron, Yulia Hicks, Peter Hall 10:05-10:30 BUBL: An Effective Region Labeling Tool Using a Hexagonal Lattice, Carolina Galleguillos, Peter Faymonville, Serge Belongie 10:30-10:55 Spectral Camera Clustering, Alexander Ladikos, Slobodan Ilic, Nassir Navab 10:55-11:15 Coffee Break (20 min) Session: Machine Learning and NN Search 11:15-12:30, 25x3 min 11:15-11:40 Discriminative Structured Outputs Prediction Model and Its Efficient Online Learning Algorithm, Yang Wu, Zejian Yuan, Yuanliu Liu, Nanning Zheng 11:40-12:05 Error-Correcting Semi-Supervised Learning with Mode-Filter on Graphs, Weiwei Du, Kiichi Urahama 12:05-12:30 Approximate Nearest Neighbor Search on HDD, Noritaka Himei, Toshikazu Wada 12:30-14:00 Lunch Break Keynote Session 14:00-15:40, 50x2 min (detail) 14:00-14:50 Keynote: Recent Advances in Image Representation for Image Segmentation, Object Class Detection and Image Classification, Frederic Jurie (University of Caen) 14:50-15:40 Keynote: Internet Multimedia: Challenges and Opportunties, Xian-Sheng Hua (Microsoft Research Asia) 15:40-16:00 Coffee Break (20 min) Session: Object Recognition 16:00-17:40, 25x4 min 16:00-16:25 Better Matching with Fewer Features: The Selection of Useful Features in Large Database Recognition Problems, Panu Turcot, David Lowe 16:25-16:50 Scaling Object Recognition: Benchmark of Current State of the Art Techniques, Mohamed Aly, Peter Welinder, Mario Munich, Pietro Perona 16:50-17:15 Robust and Efficient Recognition of Low-Quality Images by Cascaded Recognizers with Massive Local Features, Koichi Kise, Masakazu Iwamura, Kazuto Noguchi 17:15-17:40 Compressed Representation of Feature Vectors Using a Bloomier Filter and Its Application to Specific Object Recognition, Katufumi Inoue, Koichi Kise 17:40- Closing Remarks For paper presenters: Each paper is allocated a 25-minute slot for a 20-minute slide presentation and a 5-minute discussion. The following items will be available at the workshop room.
More information is available at |