Authors: Xu-Cheng Yin, Shu Tian, Ze-Yu Zuo, Wei-Yi Pei, Chun Yang

Description: Here, multi-orientation text detection is conducted with USTB_TexStar+ (USTB_TexStar [1,2] with multiple channels), and text tracking within a Bayesian framework is performed for improving detection results.
[1] Xu-Cheng Yin, Xuwang Yin, Kaizhu Huang, and Hong-Wei Hao, “Robust text detection in natural scene images”, IEEE Trans. Pattern Analysis and Machine Intelligence, 36(5): 970-983, 2014.
[2] Xu-Cheng Yin, Wei-Yi Pei, Xuwang Yin, Jun Zhang, and Hong-Wei Hao, “Multi-orientation scene text detection with adaptive clustering,” IEEE Trans. Pattern Analysis and Machine Intelligence (TPAMI), preprint, 2015.

Authors: Hyung Il Koo and Yong Gyun Kim

Description: This text detection and tracking system was submitted by Hyung Il Koo and Yong Gyun Kim from Ajou University, Suwon, Korea. The system first extracts connected components (CCs) with the maximally stable extremal region (MSER) algorithm and finds word candidates by using the bottom-up grouping clustering method [1]. For each candidate region, we find corresponding regions in the other frames with the CSK trackers [2] and build chains. We also detect collisions between the chains and filter out non-promising ones.
[1] Hyung Il Koo and Duck Hoon Kim, "Scene Text Detection via Connected Component Clustering and Nontext Filtering," IEEE Transactions on Image Processing, vol.22, no.6, pp.2296,2305, June 2013
[2] J. F. Henriques, R. Caseiro, P. Martins, J. Batista, “Exploiting the Circulant Structure of Tracking-by-detection with Kernels,”, ECCV 2012