Skip to main content

Share

Facebook Twitter LinkedIn

Center for Visual Computing Researchers Receive Marr Prize Honorable Mention at the International Conference on Computer Vision

Reseachers Presented Eight Papers at Prestigious Event
Dec. 15, 2015 -- Researchers at the UC San Diego Center for Visual Computing received a Marr Prize honorable mention for their work at the International Conference on Computer Vision (ICCV) 2015. Saining Xie and Zhuowen Tu were chosen from 1698 submissions for their paper, titled "Holistically-Nested Edge Detection."

The paper's abstract: 

We develop a new edge detection algorithm that tackles two important issues in this long-standing vision problem: (1) holistic image training and prediction; and (2) multi-scale and multi-level feature learning. Our proposed method, holistically-nested edge detection (HED), performs image-to-image prediction by means of a deep learning model that leverages fully convolutional neural networks and deeply-supervised nets. HED automatically learns rich hierarchical representations (guided by deep supervision on side responses) that are important in order to approach the human ability resolve the challenging ambiguity in edge and object boundary detection. We significantly advance the state-of-the-art on the BSD500 dataset (ODS F-score of .782) and the NYU Depth dataset (ODS F-score of .746), and do so with an improved speed (0.4 second per image) that is orders of magnitude faster than some recent CNN-based edge detection algorithms.

The paper was one of eight that faculty and students from the Center were presenting at the  ICCV, the premier forum for research in computer vision, which was held Dec. 11 to 18 in Santiago, Chile. Visual Computing Center Faculty and students also presented three papers at SIGGRAPH Asia 2015 computer graphics conference in early November.

“This level of research output in the leading computer vision and graphics fall conferences is a significant accomplishment by researchers at the newly-launched Center for Visual Computing,” said Center Director and UC San Diego computer science professor Ravi Ramamoorthi. “Our achievements this fall are an important step toward establishing UC San Diego as one of the top 5 visual computing research groups in the country.”

Center for Visual Computing papers at ICCV 2015

1. Holistically-Nested Edge Detection by Saining Xie, Zhuowen Tu


2. Learning Complexity-Aware Cascades for Deep Pedestrian Detection by Zhaowei Cai, Mohammad Saberian, Nuno Vasconcelos


3. Generic Promotion of Diffusion-Based Salient Object Detection
by Peng Jiang, Nuno Vasconcelos, Jingliang Peng


4. Bayesian Model Adaptation for Crowd Counts
by Bo Liu, Nuno Vasconcelos


5. Occlusion-aware depth estimation using light-field cameras
by Ting-Chun Wang, Alexei Efros, Ravi Ramamoorthi


6. Oriented Light-Field Windows for Scene Flow
by Pratul Srinivasan, Michael Tao, Ren Ng, Ravi Ramamoorthi


7. Photometric Stereo in a Scattering Medium
by Zak Murez, Ravi Ramamoorthi, Tali Treibitz, David Kriegman


8. Learning Concept Embeddings with Combined Human-Machine Expertise
by Michael Wilber, Iljung Kwak, David Kriegman, Serge Belongie

 

SIGGRAPH Asia 2015

Visual Computing Center Faculty and students also presented 3 papers at the recently concluded SIGGRAPH Asia 2015 conference in computer graphics, held from 2-5 November in Kobe, Japan.

1. Physically-Accurate Fur Reflectance: Modeling, Measurement and Rendering
by Ling-Qi Yan, Chi-Wei Tseng, Henrik Wann Jensen, Ravi Ramamoorthi


2. On Optimal, Minimal BRDF Sampling for Reflectance Acquisition
by Jannik Boll Nielsen, Henrik Wann Jensen, Ravi Ramamoorthi


3. Anisotropic Gaussian Mutations for Metropolis Light Transport through Hessian-Hamiltonian Dynamics
by Tzu-Mao Li, Jaakko Lehtinen, Ravi Ramamoorthi, Wenzel Jakob, Fredo Durand
 

Print News Release  Email News Release

Share
  • Facebook
  • Twitter
  • LinkedIn

Media Contacts

Ioana Patringenaru
Jacobs School of Engineering
Phone: 858-822-0899
ipatrin@ucsd.edu

RSS Feeds


Like us on Facebook Follow us on Twitter

RSS Feeds

Read our blog

Jacobs School News on YouTube

Jacobs School on Flickr