안녕하세요, BK21 공감형AI 여성공학인재양성팀에서 아래와 같이 국제 전문가 초청 세미나를 개최하오니 교수님들과 참여학생들의 많은 참여 바랍니다.
1.일시: 2025년 4월 18일 (금) 오후 2시 ~
2.장소: 온라인 zoom 세미나 (https://sookmyung-ac-kr.zoom.us/j/89625154502?pwd=yiSBZo1fbetssebzQ1i4tiU9Cs0liI.1)
3.내용:
Title: Text image analysis from low resource dataset
Abstract:
Text image understanding is a challenging research area due to the complexity and variability of text shapes. Deep learning models have shown promising results, but their effectiveness relies on large, fully labelled datasets, which are difficult to create for many scripts, especially Indic scripts. To address this, we propose an Adversarial Feature Deformation Module (AFDM) to enhance feature learning and a Diffusion-Conditioned-Diffusion Model (DCDM) for improving scene text resolution.
The AFDM learns ways to elastically warp extracted features in a scalable manner. It is inserted between intermediate layers and trained alternatively with the original framework, boosting its capability to better learn highly informative features rather than trivial ones. We record results for varying sizes of training data and observe that our enhanced network generalizes much better in the low-data regime.
The DCDM model is designed to learn the distribution of high-resolution images via two conditions: 1) the low-resolution image and 2) the character-level text embedding generated by a latent diffusion text model. The latent diffusion text module is specifically designed to generate character-level text embedding space from the latent space of low-resolution images. Additionally, the character-level CLIP module has been used to align the high-resolution character-level text embeddings with low-resolution embeddings. Our experiments on the TextZoom and Real-CE datasets demonstrate the superiority of the proposed method to state-of-the-art methods.
Zoom 호스트는 세미나 시작 10분 전 생성할 예정입니다. 참고 바랍니다.
