Sharing our vision at CVPR 2016

Published June 24, 2016

Share this page

By Andrew Fitzgibbon, Principal Researcher, Microsoft Research Cambridge

Photo by Jonathan Banks

This year, the IEEE Conference on Computer Vision and Pattern Recognition (opens in new tab) (CVPR) will take place at Caesar’s Palace from June 26–July 1 in Las Vegas, Nevada. CVPR is the premier annual computer vision event that includes the main conference and several co-located workshops and short courses. With an international roster of speakers, exhibitors, and attendees, it provides an exceptional opportunity for students, academics, and industry researchers to meet and share ideas and research results.

I am thrilled to be serving as an Area Chair along with my colleague Jian Sun, from Microsoft Research Asia. Microsoft is a Platinum Sponsor this year, with over 25 papers being presented and 40 researchers, designers, and engineers attending from across the company, representing Xbox, HoloLens, Bing, and Microsoft Research.

One new and exciting addition to CVPR this year is for industrial and academic exhibitors. The Expo, which will be open during the entire CVPR event, is a unique opportunity for multiple worlds—academics, students, budding entrepreneurs, technologists, and others—to connect and catch up on the latest news and ideas. We hope you’ll spend time at the Expo, and check out our talks, tutorials, posters, and workshops (see schedule below). Also, please stop by our booth to chat with us about projects and opportunities at Microsoft, from pedal-to-the-metal engineering to research in pure mathematics.

Conference Details

Main Conference (opens in new tab): June 27–June 30
Tutorials (opens in new tab): June 26
Workshops (opens in new tab): June 26, July 1
Industry Expo (opens in new tab): June 27–June 30

Presentations (Main Conference)

(O)	Oral
(S)	Spotlight
(P)	Poster

Session		Title
Monday
(O) Mon a.m.		The Global Patch Collider (opens in new tab)
(O) Mon a.m.		Stacked Attention Networks for Image Question Answering (opens in new tab)
(P) Mon a.m.		Efficient and Robust Color Consistency for Community Photo Collections (opens in new tab)
(P) Mon a.m.		InterActive: Inter-Layer Activeness Propagation (opens in new tab)
(P) Mon a.m.		TI-POOLING: Transformation-Invariant Pooling for Feature Learning in Convolutional Neural Networks (opens in new tab)
(P) Mon p.m.		A Multi-Level Contextual Model for Person Recognition in Photo Albums (opens in new tab)
(S) Mon p.m.		Highlight Detection with Pairwise Deep Ranking for First-Person Video Summarization (opens in new tab)
(S) Mon p.m.		You Lead, We Exceed: Labor-Free Video Concept Learning by Jointly Exploiting Web Videos and Images (opens in new tab)
Tuesday
(P) Tue a.m.		Collaborative Quantization for Cross-Modal Similarity Search (opens in new tab)
(P) Tue a.m.		Supervised Quantization for Similarity Search (opens in new tab)
(P) Tue a.m.		Efficient Intersection of Three Quadrics and Applications in Computer Vision (opens in new tab)
(P) Tue a.m.		Image Deblurring Using Smartphone Inertial Sensors (opens in new tab)
(S) Tue a.m.		Large-Scale Location Recognition and the Geometric Burstiness Problem (opens in new tab)
(S) Tue p.m.		Do It Yourself Hyperspectral Imaging with Everyday Digital Cameras (opens in new tab)
Wednesday
(O) Wed a.m.		Instance-Aware Semantic Segmentation via Multi-Task Network Cascades (opens in new tab)
(P) Wed p.m.		Joint Recovery of Dense Correspondence and Cosegmentation in Two Images (opens in new tab)
(P) Wed a.m.		Sparse to Dense 3D Reconstruction from Rolling Shutter Images (opens in new tab)
(O) Wed a.m.		ScribbleSup: Scribble-Supervised Convolutional Networks for Semantic Segmentation (opens in new tab)
Thursday
(P) Thu a.m.		MSR-VTT: A Large Video Description Dataset for Bridging Video and Language (opens in new tab)
(P) Thu a.m.		Ordinal Regression with a Multiple Output CNN for Age Estimation (opens in new tab)
(P) Thu a.m.		DisturbLabel: Regularizing CNN on the Loss Layer (opens in new tab)
(O) Thu a.m.		Jointly Modeling Embedding and Translation to Bridge Video and Language (opens in new tab)
(O) Thu p.m.		HyperDepth: Learning Depth from Structured Light Without Matching (opens in new tab)
(S) Thu p.m.		Fits Like a Glove: Fast and Easy Hand Model Personalization (opens in new tab)
(S) Thu p.m.		Semantic 3D Reconstruction with Continuous Regularization and Ray Potentials Using a Visibility Consistency Constraint (opens in new tab)