no code implementations • 30 Apr 2022 • Kerem Turgutlu, Sanat Sharma, Jayant Kumar
We demonstrate the effectiveness of our approach for object insertion in the image compositing setting and other settings like documents and design templates.
2 code implementations • Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops 2021 • Marcos V. Conde, Kerem Turgutlu
Existing computer vision research in artwork struggles with artwork's fine-grained attributes recognition and lack of curated annotated datasets due to their costly creation.
1 code implementation • 19 Jun 2021 • Marcos V. Conde, Kerem Turgutlu
In this work, we propose a multi-stage ViT framework for fine-grained image classification tasks, which localizes the informative image regions without requiring architectural changes using the inherent multi-head self-attention mechanism.