R
Ramazan Gokberk Cinbis
Researcher at Middle East Technical University
Publications - 56
Citations - 2319
Ramazan Gokberk Cinbis is an academic researcher from Middle East Technical University. The author has contributed to research in topics: Computer science & Object detection. The author has an hindex of 19, co-authored 46 publications receiving 1852 citations. Previous affiliations of Ramazan Gokberk Cinbis include French Institute for Research in Computer Science and Automation & Bilkent University.
Papers
More filters
Journal ArticleDOI
Weakly Supervised Object Localization with Multi-Fold Multiple Instance Learning
TL;DR: This work follows a multiple-instance learning approach that iteratively trains the detector and infers the object locations in the positive training images and proposes a window refinement method, which improves the localization accuracy by incorporating an objectness prior.
Proceedings ArticleDOI
Multi-fold MIL Training for Weakly Supervised Object Localization
TL;DR: This work follows a multiple-instance learning approach that iteratively trains the detector and infers the object locations in the positive training images, which improves detection performance and prevents training from prematurely locking onto erroneous object locations.
Proceedings ArticleDOI
Unsupervised metric learning for face identification in TV video
TL;DR: This paper addresses the identification problem for face-tracks that are automatically collected from uncontrolled TV video data and learns a cast-specific metric, adapted to the people appearing in a particular video, without using any supervision.
Proceedings ArticleDOI
Gradient Matching Generative Networks for Zero-Shot Learning
TL;DR: This work proposes a generative model that can naturally learn from unsupervised examples, and synthesize training examples for unseen classes purely based on their class embeddings, and therefore, reduce the zero-shot learning problem into a supervised classification task.
Proceedings ArticleDOI
Segmentation Driven Object Detection with Fisher Vectors
TL;DR: A method to produce tentative object segmentation masks to suppress background clutter in the features to improve object detection significantly and exploit contextual features in the form of a full-image FV descriptor, and an inter-category rescoring mechanism.