Jiawei He (PhD), Zhiwei Deng (PhD), and Moustafa S. Ibrahim (PhD) had a paper accepted to IEEE Winter Conf. on Applications of Computer Vision (WACV), 2018. A class-independent action tublet network for human action localization is presented.
	Akash Abdu Jyothi successfully defended his M.Sc. thesis Generating Natural Language Summaries for Image Sets. Congratulations Akash!
	Nelson Nauata successfully defended his M.Sc. thesis Structured Label Inference for Visual Understanding. Congratulations Nelson!
	Jon Smith successfully defended his M.Sc. thesis REP3D: 3D Human Motion Capture Dataset for Athletic Movement. Congratulations Jon!
	Xiaoyu Liu successfully defended her M.Sc. thesis Joint Constrained Clustering and Feature Learning based on Deep Neural Networks. Congratulations Xiaoyu!
	Lei Chen (PhD) and Mengyao Zhai (PhD) had a paper accepted to the 5th Workshop on Web-scale Vision and Social Media (VSM), 2017. A weakly-supervised approach to learning action models from multi-person video is presented.
	Fred Tung (PDF) and Srikanth Muralidharan (PhD) had a paper accepted to the British Machine Vision Conference (BMVC), 2017. Deep network fine-tuning and pruning are jointly conducted in a Bayesian optimization framework.
	Nazanin Mehrasa successfully defended her M.Sc. thesis Learning Person Trajectory Representations for Team Activity Analysis. Congratulations Nazanin!
	Karoon Rashedi Nia successfully defended his M.Sc. thesis Automatic Building Damage Assessment Using Deep Learning and Ground-Level Image Data. Congratulations Karoon!
	Yatao Zhong successfully defended his M.Sc. thesis Learning Person Trajectory Features for Sports Video Analysis. Congratulations Yatao!
	Mehran Khodabandeh (PhD), Srikanth Muralidharan (PhD), and Nazanin Mehrasa (MSc) had a paper accepted to IAPR International Conference on Pattern Recognition (ICPR), 2016. An unsupervised method for learning features for video segmentation is presented.
	Srikanth Muralidharan successfully defended his M.Sc. thesis A Hierarchical Deep Temporal Model for Group Activity Recognition. Congratulations Srikanth!
	Zhiwei Deng (PhD), Arash Vahdat (URA), and Hexiang Hu (BSc) had a paper accepted to IEEE Computer Vision and Pattern Recognition (CVPR), 2016. Structure learning for a message passing deep network is developed and applied to group activity recognition.
	Hexiang Hu (BSc) and Zhiwei Deng (PhD) had a paper accepted to IEEE Computer Vision and Pattern Recognition (CVPR), 2016. Images are tagged with labels from a hierarchy via a deep learning algorithm that reasons about label relations.
	Moustafa S. Ibrahim (PhD), Srikanth Muralidharan (MSc), Zhiwei Deng (PhD), and Arash Vahdat (URA) had a paper accepted to IEEE Computer Vision and Pattern Recognition (CVPR), 2016. A multi-level temporal deep network for recognizing group activities in videos is developed.
	Lei Chen successfully defended his M.Sc. thesis Learning Action Primitives for Multi-Level Video Event Understanding. Congratulations Lei!
	Zhiwei Deng successfully defended his M.Sc. thesis Deep Structured Models for Group Activity Recognition. Congratulations Zhiwei!
	Mengyao Zhai successfully defended her M.Sc. thesis Object Detection in Surveillance Video from Dense Trajectories. Congratulations Mengyao!
	Guang-Tong Zhou successfully defended his Ph.D. thesis Toward Scene Recognition by Discovering Semantic Structures and Parts. Congratulations Guang-Tong!
	Hossein Hajimirsadeghi successfully defended his Ph.D. thesis Multiple Instance Learning for Visual Recognition: Learning Latent Probabilistic Models. Congratulations Hossein!
	Hossein Hajimirsadeghi (PhD) had a paper accepted to IEEE International Conference on Computer Vision (ICCV), 2015. The paper presents a boosting method for combining multiple high-capacity models in a structured prediction setting, demonstrating results on a variety of applications including group activity recognition and 3d action recognition.
	Zhiwei Deng (MSc), Mengyao Zhai (MSc), Lei Chen (MSc), Yuhao Liu (BSc), and Srikanth Muralidharan (MSc) had a paper accepted to the British Machine Vision Conference (BMVC), 2015. A deep learning framework for group activity recognition via message passing is presented.
	Wang Yan (PDF) and Jordan Yap (BSc) had a paper accepted to the British Machine Vision Conference (BMVC), 2015. Video retrieval in a "one-shot" (single example) setting is improved by leveraging multi-task learning from existing models.
	Mehran Khodabandeh (PhD), Hossein Hajimirsadeghi (PhD), Arash Vahdat (URA), and Guang-Tong Zhou (PhD) had a paper accepted to Workshop on Group and Crowd Behavior Analysis and Understanding at CVPR. Human interaction discovery in surveillance video is addressed using a human-in-the-loop clustering framework.
	Hossein Hajimirsadeghi (PhD), Arash Vahdat (URA), and Wang Yan (PDF) had a paper accepted to IEEE Computer Vision and Pattern Recognition (CVPR), 2015. The paper presents a method for multiple instance learning with cardinality potential kernels, with applications to group activity recognition, video summarization, and internet video event detection.
	Yasaman Sefidgar (MSc) and Arash Vahdat (URA) had a paper accepted to Computer Vision and Image Understanding (CVIU). Human-human and human-vehicle interactions are detected in surveillance video using a discriminative model.
	Mengyao Zhai (MSc), Lei Chen (MSc), Mehran Khodabandeh (PhD), and Jinling Li, had a paper accepted to IAPR Machine Vision Applications (MVA), 2015. Dense trajectory features are used in a regression model for vehicle and person detection in videos.
	Mehran Khodabandeh, successfully defended his M.Sc. thesis Discovering Human Interactions in Videos with Limited Data Labeling, Congratulations Mehran!
	Arash Vahdat, successfully defended his Ph.D. thesis Weakly Supervised Models For Recognizing And Clustering High-Level Complex Events In Video. Congratulations Arash!
	Nataliya Shapovalova successfully defended her Ph.D. thesis Towards Action Recognition and Localization in Videos with Weakly Supervised Learning. Congratulations Nataliya!
	Jinling Li successfully defended her M.Sc. thesis Road User Detection and Analysis in Traffic Surveillance Videos. Congratulations Jinling!
	Arash Vahdat (PhD) and Guang-Tong Zhou (PhD) had a paper accepted to the European Conference on Computer Vision (ECCV), 2014. An approach for discovering clusters of related internet videos is presented, utilizing noisy tag labels and image features.
	Yasaman Sefidgar successfully defended her M.Sc. thesis Discriminative Key-Segment Model for Interaction Detection. Congratulations Yasaman!
	Amir Bakhtiari successfully defended his M.Sc. thesis Detecting Pedestrians Using Motion Patterns: A Latent Tracking Approach. Congratulations Amir!
	Nataliya Shapovalova (PhD) had a paper accepted to Neural Information Processing Systems (NIPS), 2013. A joint model for visual saliency and action recognition in videos is described, trained in a latent SVM framework utilizing eye gaze data.
	Guang-Tong Zhou (PhD), Tian Lan (PhD), and Arash Vahdat (PhD) had a paper accepted to Neural Information Processing Systems (NIPS), 2013. The paper describes an approach for clustering with latent variables under a max-margin criterion, with application to videos of human activities.
	Tian Lan successfully defended his Ph.D. thesis From Flat to Hierarchical: Modeling Structures in Visual Recognition. Congratulations Tian!
	Arash Vahdat (PhD) had a paper accepted to IEEE International Conference on Computer Vision (ICCV), 2013. An algorithm for predicting a set of tags that describe and image or video is described, trained from noisy tag annotations.
	Arash Vahdat (PhD) and Kevin Cannons (PDF) had a paper accepted to IEEE International Conference on Computer Vision (ICCV), 2013. A compositional model for video event recognition is presented, based on a novel multiple kernel learning algorithm that incorporates latent variables.
	Tian Lan (PhD) had a paper accepted to IEEE International Conference on Computer Vision (ICCV), 2013. A multi-level object detection framework that models sub-categories through to visual composites is developed.
	Hossein Hajimirsadeghi (PhD) and Jinling Li (MSc) had a paper accepted to the Conference on Uncertainty in Artificial Intelligence (UAI), 2013. A multiple instance learning algorithm is developed, which models varying levels of positive instances in a bag. A discriminative training algorithm is proposed, based on efficient inference of cardinality-based clique potential functions.
	Tian Lan (PhD) had a paper accepted to IEEE Computer Vision and Pattern Recognition (CVPR), 2013. This work develops an image tag ranking algorithm that can determine which keywords are more relevant to a given image.
	Guang-Tong Zhou (PhD), Tian Lan (PhD), and Weilong Yang had a paper accepted to IEEE Computer Vision and Pattern Recognition (CVPR), 2013. An image classification algorithm that matches a set of objects to an image is developed.
	Yuke Zhu (BSc) and Tian Lan (PhD) had a paper on activity recognition accepted to IAPR Conference on Machine Vision Applications (MVA), 2013. Latent variable models were used to analyze nursing home surveillance video.
	Weilong Yang successfully defended his Ph.D. thesis Discriminative Latent Variable Models For Visual Recognition. Congratulations Weilong!
	Weilong Yang (PhD), and Arash Vahdat (PhD) had a paper accepted to Neural Information Processing Systems (NIPS) 2012. A kernelized latent SVM formulation is developed, and applied to various visual recognition problems.
	Zhi Feng Huang successfully defended his M.Sc. thesis Latent Boosting For Action Recognition. Congratulations Zhi Feng!
	Hossein Hajimirsadeghi (PhD) had a paper accepted to the IAPR International Conference on Pattern Recognition (ICPR), 2012. A boosting-based multiple instance learning algorithm is developed. The algorithm allows for varying degrees of bag-level positives in the multiple instance learning setup.
	Nataliya Shapovalova (PhD), Arash Vahdat (PhD), Tian Lan (PhD) and Kevin Cannons (PDF) had a paper accepted to the European Conference on Computer Vision (ECCV), 2012. The work focuses on weakly-supervised action recognition in videos. We learn a model that can classify unseen test videos, as well as localize a region of interest in the video that captures the discriminative essence of the action class.
	Tian Lan (PhD) and Weilong Yang (PhD) had a paper accepted to the European Conference on Computer Vision (ECCV), 2012. An image retrieval approach is developed. The approach allows queries specifying which objects, in which relations, should be present in an image. A learning algorithm for determining which object detectors and relations are most useful for retrieval is presented.
	Brian Milligan successfully defended his M.Sc. thesis Selecting And Commanding Groups Of Robots Using A Vision-Based Natural User Interface. Congratulations Brian!
	Mani Ranjbar successfully defended his Ph.D. thesis Optimizing Non-Decomposable Loss Functions In Structured Prediction. Congratulations Mani!
	Tian Lan (PhD) had a paper accepted to IEEE Computer Vision and Pattern Recognition (CVPR), 2012. This work addresses human activity recognition. A model combining low-level actions, high-level events, and social roles of people was developed.
	Mani Ranjbar (PhD) and Arash Vahdat (PhD) had a paper accepted to IEEE Computer Vision and Pattern Recognition (CVPR), 2012. An optimization strategy using dual decomoposition for learning with complex loss functions is described.
	Tian Lan (PhD), Weilong Yang (PhD), and Yang Wang had a paper accepted to IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI). Methods for modeling context relating the actions of individuals and groups of people are compared and combined (earlier NIPS 2010 and SGA 2010 papers).
	Arash Vahdat (MSc), Bo Gao (MSc), and Mani Ranjbar (PhD) had a paper accepted to the Eleventh IEEE International Workshop on Visual Surveillance (VS), 2011. A keypose-based model for recognizing human interactions in videos is presented.
	Pengfei Yu successfully defended his M.Sc. thesis Image Classification Using Latent Spatial Pyramid Matching. Congratulations Pengfei!
	Tian Lan (PhD) had a paper accepted to the IEEE International Conference on Computer Vision (ICCV), 2011. A discriminative model for jointly recognizing and localizing actions is presented.
	Zhi Feng Huang (MSc) and Weilong Yang (PhD) had a paper accepted to the 22nd British Machine Vision Conference (BMVC), 2011. A boosting algorithm for models with latent variables is developed, and applied to action recognition.
	Bo Gao sucessfully defended his M.Sc. thesis Exemplar-Based Human Interaction Recognition: Features And Key Pose Sequence Model. Congratulations Bo!
	Arash Vahdat successfully defended his M.Sc. thesis A Key Pose Model For Human Interaction Recognition And Color From Gray By Optimized Color Ordering. Congratulations Arash!
	Brian Milligan won the best video award at the 6th ACM/IEEE International Conference on Human-Robot Interaction (HRI) 2011 for his video titled Selecting and Commanding Groups in a Multi-Robot Vision Based System.
	Ferdinand Stefanus successfully defended his M.Sc. thesis Automatic Pedestrian Detection and Tracking with a Multiple-Cue Max-Margin Framework. Congratulations Ferdinand!
	Tian Lan (MSc), Weilong Yang (PhD), and Yang Wang had a paper accepted to Neural Information Processing Systems (NIPS) 2010. Context relating the actions of individuals and groups of people is modelled for human action recognition. The paper develops a latent variable framework that adapts the latent connections between actions of individuals to model an image.
	Yang Wang had a paper accepted to Neural Information Processing Systems (NIPS) 2010. The paper develops a method for annotating images with tags describing the objects present. The model explicitly captures the matching of object tags to image regions in a discriminative framework.
	Tian Lan successfully defended his M.Sc. thesis Beyond Actions: Discriminative Models For Contextual Group Activities. Congratulations Tian!
	Bahman Yari Saeed Khanloo successfully defended his M.Sc. thesis Combining Simple Trackers Using Structural SVMs For Offline Single Object Tracking. Congratulations Bahman!
	Tian Lan (MSc) and Yang Wang had a paper on action recognition accepted in the International Workshop on Sign, Gesture, Action, 2010. A contextual feature representation that considers the actions of nearby people when recognizing the action of an individual is presented. For example, one can recognize that a nursing home resident has fallen based on the fact that other people in the scene are coming to help.
	Yang Wang had a paper on object recognition accepted in the European Conference on Computer Vision (ECCV), 2010. Object categories (e.g. bird, apple, chair) and attributes (e.g. red, shiny, edible) are treated together in a unified learning framework.
	Mani Ranjbar (PhD) and Yang Wang had a paper on object segmentation accepted in the European Conference on Computer Vision (ECCV), 2010. A learning framework for choosing segmentation algorithm parameters against a class of complex loss measures is developed.
	Weilong Yang successfully defended his M.Sc. thesis Learning Transferable Distance Functions For Human Action Recognition And Detection. Congratulations Weilong!
	Zhi Feng Huang (MSc) won an NSERC Alexander Graham Bell Canada Graduate Scholarship (CGS). Congratulations!
	Weilong Yang (MSc) and Yang Wang had a paper on action recognition in still images accepted in the IEEE Computer Vision and Pattern Recognition (CVPR), 2010. Body pose is treated as a latent variable in a model focused on action recognition.
	Bahman Yari Saeed Khanloo (MSc), Ferdinand Stefanus (MSc), and Mani Ranjbar (PhD) had a paper on tracking accepted in the Seventh Canadian Conference on Computer and Robot Vision (CRV), 2010. A max-margin criterion is used to learn weights for parameters of a tracking algorithm with multiple features.
	Alex Couture-Beil (MSc) developed an algorithm for selecting individual robots using face detection and commanding them using gesture recognition. A video of the system appeared at the IEEE International Conference on Human-Robot Interaction (Video Session) (HRI), 2010. A full paper describing the work will appear at the Seventh Canadian Conference on Computer and Robot Vision (CRV), 2010.
	Mohammad Norouzi successfully defended his M.Sc. thesis Convolutional Restricted Boltzmann Machines for Feature Learning. Congratulations Mohammad!
	Yang Wang had a paper on semi-supervised learning for structured prediction problems accepted to Advances in Neural Information Processing Systems (NIPS), 2009. A novel information theoretic approach to mutual information-based regularization is developed.
	Dr. Yang Wang successfully defended his Ph.D. thesis Learning Structured Models For Human Actions And Poses. Congratulations Yang!
	Weilong Yang (MSc) and Yang Wang (PhD) had papers on transfer learning for human action recognition accepted to the Asian Conference on Computer Vision (ACCV) and 2nd International Workshop on Machine Learning for Vision-based Motion Analysis (MLVMA at ICCV), 2009. Video clips are compared with a distance function built from patches. The weights of these patches are determined via transfer learning (MLVMA), and used in an efficient action detection framework (ACCV).
	William Ma successfully defended his M.Sc. thesis Motion Estimation For Functional Medical Imaging Studies Using A Stereo Video Head Pose Tracking System. Congratulations William!
	Yang Wang (PhD) had a paper on human action recognition accepted to IEEE Computer Vision and Pattern Recognition (CVPR), 2009. A hidden conditional random field model is learned using a max-margin criterion.
	Mohammad Norouzi (MSc) and Mani Ranjbar (PhD) had a paper on feature learning accepted to IEEE Computer Vision and Pattern Recognition (CVPR), 2009. A hierarchical convolutional variant of the restricted Boltzmann machine model is developed in which parameters are tied to learn filters which are used as features for object detection.
	Mark Bayazit (BSc) and Alex Couture-Beil (MSc) had a paper on human gesture recognition accepted to IAPR Conference on Machine Vision Applications (MVA), 2009. An efficient algorithm using the GPU was developed.
	Yang Wang (PhD) had a paper on human action recognition accepted in IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI), 2009. A novel bag-of-words sequence model is developed, based on latent Dirichlet allocation (LDA).
	Yang Wang (PhD) had a paper on human action recognition accepted to Advances in Neural Information Processing Systems (NIPS), 2008. A hidden conditional random field model incorporating large-scale global features and local patch features is used in this work.
	Yang Wang (PhD) had a paper on estimating human pose. Multiple tree-structured models are combined to handle occlusion and avoid "double-counting" of image evidence. This work appeared at the European Conference on Computer Vision (ECCV), 2008.
	Bo Chen (BSc) and William Ma (MSc) had a paper on energy efficient scheduling for window-scanning algorithms at the Workshop on the Interaction between Operating Systems and Computer Architecture (WIOSCA), 2008.
	Yang Wang (PhD) had a paper on boosting with incomplete information. For example, in object recognition bounding box information may be provided on some training data. This work appeared at the International Conference on Machine Learning (ICML), 2008.