โก LLM Efficiency¶
๐๏ธ ECCV2026 ยท 1 paper notes
๐ Same area in other venues: ๐ท CVPR2026 (8) ยท ๐ฌ ICLR2026 (171) ยท ๐ฌ ACL2026 (23) ยท ๐งช ICML2026 (48) ยท ๐ค AAAI2026 (9) ยท ๐ง NeurIPS2025 (34)
- ExpoMotion: A Large-Scale Benchmark and A Householder Projection Network for Multi-Exposure Fusion
-
To tackle the lack of high-fidelity dynamic benchmarks and the propensity of correlation-based fusion to admit ghosting artifacts, this paper presents ExpoMotion, a 1,738-sequence real-world dynamic MEF benchmark, and HOP, a decoupled architecture using Householder orthogonal projections to filter out motion artifacts geometrically while harmonizing global exposures.