F
Fatih Porikli
Researcher at Australian National University
Publications - 58
Citations - 2710
Fatih Porikli is an academic researcher from Australian National University. The author has contributed to research in topics: Deep learning & Discriminative model. The author has an hindex of 21, co-authored 58 publications receiving 1891 citations. Previous affiliations of Fatih Porikli include Commonwealth Scientific and Industrial Research Organisation & NICTA.
Papers
More filters
Proceedings ArticleDOI
See More, Know More: Unsupervised Video Object Segmentation With Co-Attention Siamese Networks
TL;DR: In this paper, a co-attention Siamese network (COSNet) is proposed to address the unsupervised video object segmentation task from a holistic view.
Posted Content
See More, Know More: Unsupervised Video Object Segmentation with Co-Attention Siamese Networks
TL;DR: This work introduces a novel network, called as CO-attention Siamese Network (COSNet), to address the unsupervised video object segmentation task from a holistic view and proposes a unified and end-to-end trainable framework where different co-att attention variants can be derived for mining the rich context within videos.
Proceedings ArticleDOI
Beyond Local Search: Tracking Objects Everywhere with Instance-Specific Proposals
TL;DR: In this paper, an instance-specific objectness measure is used to generate a small number of "high-quality" proposals and evaluate them against the object model that can be adopted from an existing tracking-by-detection approach as a core tracker.
Journal ArticleDOI
Quadruplet Network With One-Shot Learning for Fast Visual Object Tracking
TL;DR: In this paper, a quadruplet deep network is proposed to examine the potential connections among the training instances, aiming to achieve a more powerful representation, and a new weight layer is introduced to automatically select suitable combination weights, which will avoid the conflict between triplet and pair loss leading to worse performance.
Book ChapterDOI
Zero-Shot Object Detection: Learning to Simultaneously Recognize and Localize Novel Concepts
TL;DR: This work introduces a new Zero-Shot Detection problem setting, which aims at simultaneously recognizing and locating object instances belonging to novel categories without any training examples, and designs an original loss function that achieves synergy between max-margin class separation and semantic space clustering.