Company News

Fast object detection using boosted co-occurrence histograms of oriented gradients

Haoyu Ren, Cher Keng Heng, Wei Zheng, Luhong Liang, Xilin Chen
International Conference on Image Processing (ICIP) 2010, 2010.12

Co-occurrence histograms of oriented gradients (CoHOG) are powerful descriptors in object detection. In this paper, we propose to utilize a very large pool of CoHOG features with variable-location and variable-size blocks to capture salient characteristics of the object structure. We consider a CoHOG feature as a block with a special pattern described by the offset. A boosting algorithm is further introduced to select the appropriate locations and offsets to construct an efficient and accurate cascade classifier. Experimental results on public datasets show that our approach simultaneously achieves high accuracy and fast speed on both pedestrian detection and car detection tasks.

Link: https://ieeexplore.ieee.org/document/5651963

Back

Share

ISDN Holdings’ NovaPeak partners with Panasonic HD to Scale LiveInspect.AI globally

LiveInspect.AI Development Team lead by Dr. Takeshi Ando Managing Director of Panasonic R&D Center Singapore (front right) and Mr. Eugene…

Panasonic launched a “Panasonic Deep Tech Innovation Challenge” inviting local start-ups to leverage Panasonic’s technology & patents for business creation

Panasonic R&D Center Singapore has launched a Panasonic Deep Tech Innovation Challenge organized by Panasonic, ICMG and ACE. This Challenge…

Paper accepted for publication in the IEEE Transactions on Systems, Man, and Cybernetics: Systems

Collaborating with Nanjing University of Science and Technology (NUST), Panasonic R&D Center Singapore co-authored a paper titled “Modular Lightweight Network…

Panasonic R&D Center Singapore Achieving the World’s Top Best Accuracy for Face Recognition

In collaboration with the National University of Singapore (NUS) and the Innovation Center of Connected Solutions of Panasonic Corporation, Panasonic R&D Center Singapore has recently developed face recognition technology that has achieved the world’s Top Best Accuracy for Face Recognition by benchmarking the dataset covering all the conditions taken in the surveillance market, the dataset of which is provided by the National Institute of Standards and Technology (NIST).