9 papers
Learning Structural Illumination for Unsupervised Low-light Enhancement
Tianle Du, Peiyuan He, Hainuo Wang +2
Existing unsupervised low-light image enhancement (LLIE) methods often estimate illumination directly from the entire low-light input, without separating its spatially varying illu…
Internally Referenced Low-Light Enhancement
Peiyuan He, Hainuo Wang, Hengxing Liu +2
Self-supervised low-light image enhancement (LLIE) is highly appealing as it eliminates the reliance on external paired data. However, the lack of external references causes networ…
Representative Attention For Vision Transformers
Yuntong Li, Hainuo Wang, Hengxing Liu +2
Linear attention has emerged as a promising direction for scaling Vision Transformers beyond the quadratic cost of dense self-attention. A prevalent strategy is to compress spatial…
On the Global Photometric Alignment for Low-Level Vision
Mingjia Li, Tianle Du, Hainuo Wang +2
Supervised low-level vision models rely on pixel-wise losses against paired references, yet paired training sets exhibit per-pair photometric inconsistency, say, different image pa…
WiT: Waypoint Diffusion Transformers via Trajectory Conflict Navigation
Hainuo Wang, Mingjia Li, Xiaojie Guo
While recent Flow Matching models avoid the reconstruction bottlenecks of latent autoencoders by operating directly in pixel space, the lack of semantic continuity in the pixel man…
Anchor then Polish for Low-light Enhancement
Tianle Du, Mingjia Li, Hainuo Wang +1
Low-light image enhancement is challenging due to entangled degradations, mainly including poor illumination, color shifts, and texture interference. Existing methods often rely on…