cs.CVJul 11, 2026

BOCCHI: A More Realistic and Challenging Benchmark for Local Motion Blur Detection with MSDCT-UNet

Authors: Kuan-Lin Chen, Yuan-Kang Lee, Cheng-Yuan Chiang, Jian-Jiun Ding

Organizations: Graduate Institute of Communication Engineering, National Taiwan University, Taipei, Taiwan · MediaTek Inc., Hsinchu, Taiwan

Abstract

Local motion blur detection requires pixel-level localization of blurred regions. Existing benchmarks let models rely on gradient shortcuts that fail to transfer. We introduce BOCCHI (Blurred Objects Captured across Cameras with Human-annotated Imagery), a real-captured benchmark whose sharp regions overlap the blur gradient distribution and defeat these shortcuts, and propose MSDCT-UNet (Multi-Scale Discrete Cosine Transform UNet), a frequency-aware encoder-decoder injecting multi-scale DCT priors through DCT Attention and FiLM. MSDCT-UNet ranks first in in-domain mIoU and boundary localization on BOCCHI, and BOCCHI-trained models outperform every other training source on cross-dataset transfer with only 633 training images.

Explore similar work

CardsList
  1. SloMoDeblur: A Large-Scale Smartphone Image Deblurring Dataset

    Jun 24, 2025Syed Mumtahin Mahmud, Mahdi Mohd Hossain Noki, Prothito Shovon Majumder +4Motion BlurSmartphone Photographs