Skip to content
View AaronCIH's full-sized avatar
14 followers · 2 following

Achievements

Achievement: Starstruck

Achievements

Achievement: Starstruck

Block or report AaronCIH

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
AaronCIH/README.md

I-Hsiang (Aaron) Chen

Computer Vision · Generative AI · Multimodal Learning

Portfolio · Publications · Experience · Awards · CV

Bridging vision, generation, and the real world — Aaron Chen's research portfolio

About me

I explore how machines perceive, restore, and understand the visual world. My research connects generative models, robust visual learning, and multimodal understanding, with a focus on systems that work beyond clean, familiar conditions.

My research background spans National Taiwan University and the University of Washington, alongside industry research experience at Samsung Research, MediaTek, and ASML.

Research interests

  • Image Generation and Editing — diffusion priors, image restoration, and task-aware visual generation.
  • Multimodal Learning — connecting visual understanding, generation, and retrieval.
  • Domain Generalization — learning representations that remain useful in unfamiliar environments.
  • Visual Understanding — semantic segmentation, crowd counting, and re-identification.

Selected research

Preview Project
RAR teaser showing iterative image quality assessment and restoration Restore, Assess, Repeat (RAR)
CVPR 2026 · First author

A unified framework that restores an image, assesses its quality, and repeats the process to handle unknown and composite degradations.

Project · Paper · Video
RobustVisRAG visual document retrieval comparison RobustVisRAG
CVPR 2026 · First author

Causality-aware visual retrieval and grounded answer generation from documents affected by blur, noise, and other visual degradations.

Project · Paper · Video
UniRestore teaser comparing perceptual and task-oriented image restoration UniRestore
CVPR 2025 Highlight · First author

Bridging perceptual quality and downstream task needs through a unified image restoration model using a diffusion prior.

Project · Paper · Video
PDAF semantic segmentation comparison on an urban scene PDAF
ICCV 2025 · First author

Probabilistic diffusion alignment and latent domain modeling for semantic segmentation beyond familiar conditions.

Project · Paper · Video
APGCC crowd localization example with point detections APGCC
ECCV 2024 · First author

Auxiliary Point Guidance for more stable point-based crowd counting and localization.

Project · Paper · Video

Explore the full research collection and animated previews →

Research & industry experience

Organization Focus
University of Washington Visiting research on diffusion models and image enhancement
National Taiwan University Graduate research in image enhancement, re-identification, and crowd counting
Samsung Research UK Unified multimodal understanding and generation
MediaTek End-to-end learning-based video compression
ASML Synthetic data generation for defect detection

More about my research journey →

Selected recognition

  • NTU Outstanding Young Award — 2025
  • CTCI Research Award — 2023
  • Hon-Hai Tech Award — 2022

All honors and awards →

Let's connect

I'm happy to connect about research and collaboration in computer vision, generative AI, and multimodal learning.

Email · LinkedIn · Personal website

Pinned Loading

  1. Awesome-AutoSkill-AutoRubric Awesome-AutoSkill-AutoRubric Public

    A curated collection of papers & repos on Auto-Skill (self-evolving agents) and Auto-Rubric (rubric learning from preferences) for LLM alignment & customization.

    7 1

  2. robustvisrag/RobustVisRAG robustvisrag/RobustVisRAG Public

    CVPR26 - RobustVisRAG: Causality-Aware Vision-Based Retrieval-Augmented Generation under Visual Degradations

    5

  3. Restore-Assess-Repeat/Restore-Assess-Repeat.github.io Restore-Assess-Repeat/Restore-Assess-Repeat.github.io Public

    CVPR2026-Restore, Assess, Repeat: A Unified Framework for Iterative Image Restoration

    JavaScript 1

  4. APGCC APGCC Public

    ECCV24 - Improving Point-based Crowd Counting and Localization Based on Auxiliary Point Guidance

    Python 97 27

  5. unirestore/UniRestore unirestore/UniRestore Public

    CVPR25(Highlight)-Unified Perceptual and Task-Oriented Image Restoration Model Using Diffusion Prior

    Python 100 5

  6. pdaf-iccv/PDAF pdaf-iccv/PDAF Public

    Python 8 1