Publications

Multi-Concept T2I-Zero: Tweaking Only The Text Embeddings and Nothing Else

Multi-Concept T2I-Zero: Tweaking Only The Text Embeddings and Nothing Else

Hazarapet Tunanyan, Dejia Xu, Shant Navasardyan, Zhangyang Wang, and Humphrey Shi

Arxiv preprint, 2023

In this work, we consider a more ambitious goal: natural multi-concept generation using a pre-trained diffusion model, and with almost no extra cost.

arXiv Project

Specialist Diffusion: Plug-and-Play Sample-Efficient Fine-Tuning of Text-to-Image Diffusion Models to Learn Any Unseen Style

Specialist Diffusion: Plug-and-Play Sample-Efficient Fine-Tuning of Text-to-Image Diffusion Models to Learn Any Unseen Style

Haoming Lu, Hazarapet Tunanyan, Kai Wang, Shant Navasardyan, Zhangyang Wang, and Humphrey Shi

IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR), 2023

In this paper, we aim to learn an unseen style, so that the fine-tuned model can generate high-quality images of arbitrary objects in this style.

PDF Code Project Poster

C4Net: Contextual Compression and Complementary Combination Network for Salient Object Detection

C4Net: Contextual Compression and Complementary Combination Network for Salient Object Detection

Hazarapet Tunanyan

BMVC, 2021

The paper proposes a salient object detection model using feature concatenation, complementary feature extraction, excessiveness loss, and pyramid-semantic guidance.

arXiv