Bio-SFT: Asymmetric Cortical Guidance and Retinal Adaptation for Robust HDR Reconstruction
2026-07-20 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors created a new method called Bio-SFT to turn ordinary photos with limited brightness into images showing much wider brightness ranges, like HDR photos. They used ideas inspired by how the human eye and brain process light and detail, such as a model mimicking eye adaptation and splitting information into types like the Parvo and Magno pathways. Their approach also uses a special event-driven step to reduce noise in dark areas while keeping details clear. Tests showed their method improves image quality and reduces errors compared to other techniques.
HDR reconstructionSDR imageNaka-Rushton adaptationParvo and Magno pathwaysSpiking Neural Network (SNN)TransformerEvent-driven gatingNoise suppressionPerceptual image qualityHDRTV1K dataset
Authors
Tingyu Cheng, Ting Zhang, Chongyi Li, Zhaoqing Pan, Tiesong Zhao
Abstract
Recovering high dynamic range (HDR) radiance from a single standard dynamic range (SDR) image is highly ill-posed. Extreme luminance variation and severe quantization in dark regions make accurate reconstruction challenging, often leading to visual artifacts and color distortions. To address this problem, we propose Bio-SFT, a bio-inspired spiking frequency transformer for single-image HDR reconstruction. Bio-SFT incorporates three biologically motivated components. First, a learnable Naka--Rushton retinal adaptation frontend stabilizes the input under complex lighting conditions. Second, an explicit Parvo--Magno split introduces asymmetric Parvo-to-Magno guidance, allowing high-frequency structural cues to modulate low-frequency reconstruction. Third, an event-driven SNN hard gating module applies all-or-none spiking to suppress dark-region noise while preserving structural details. The module is trained with a sparsity prior to encourage efficient feature utilization. Built for end-to-end training within a transformer backbone, these lightweight components provide strong parameter efficiency. Experiments on HDRTV1K show that Bio-SFT achieves competitive perceptual quality and consistently improves HDR-VDP-3 and $ΔE_{ITP}$ while reducing artifact propagation in symmetric guidance pipelines.