Publications by authors named "Ramazan Gokberk Cinbis"
Article Synopsis
- This paper addresses zero-shot sign language recognition (ZSSLR), aiming to identify unseen sign classes using previously learned models from seen classes.
- It utilizes textual descriptions and attributes from sign language dictionaries as semantic representations for knowledge transfer in the recognition process.
- The authors introduce three benchmark datasets and demonstrate how combining these textual and visual representations can effectively recognize new sign classes, opening avenues for further research in this area.
Weakly Supervised Object Localization with Multi-Fold Multiple Instance Learning.
IEEE Trans Pattern Anal Mach Intell
January 2017
Object category localization is a challenging problem in computer vision. Standard supervised training requires bounding box annotations of object instances. This time-consuming annotation process is sidestepped in weakly supervised learning.
View Article and Find Full Text PDFApproximate Fisher Kernels of Non-iid Image Models for Image Categorization.
IEEE Trans Pattern Anal Mach Intell
June 2016
The bag-of-words (BoW) model treats images as sets of local descriptors and represents them by visual word histograms. The Fisher vector (FV) representation extends BoW, by considering the first and second order statistics of local descriptors. In both representations local descriptors are assumed to be identically and independently distributed (iid), which is a poor assumption from a modeling perspective.
View Article and Find Full Text PDF