In this context, a zero-reference image enhancement network for low light conditions is proposed in this paper. First, the improved Encoder-Decoder structure is used to extract image features to generate feature maps and generate the parameter matrix of the enhancement factor from the feature maps...
在本文中,我们提出了一种新的零参考低照度增强框架,该框架可仅使用正常光照下的图像进行训练。为此,我们从物理光传递理论中汲取灵感,设计了一种光照不变先验。这个先验值是连接正常光线图像和弱光图像的桥梁。然后,我们开发了一个先验图像框架,在没有弱光数据的情况下进行训练。在测试过程中,该框架能够将我们的光照...
For low-light enhancement tasks, network architecture design needs to be adapted to the characteristic of low-light images having more low-light features than high-light features. At the same time, for low-light enhancement tasks in real scenes, zero-shot learning32 methods are needed to ...
首先使用CLIP图像编码器Φimg和使用CLIP文本编码器Φtxt学习的提示对增强图像ˆI进行编码。接下来,计算嵌入增强图像 Φimg ( ^I) 和学习提示对 Φtxt(P ) 之间的余弦相似度,其中 P = Pp, Pn 是一对正负学习提示。 (2)语义指导 利用CLIP 模型的 zeroshot 能力,以直接的方式在训练期间引入语义指导。为了简...
(2019), propose a zero-shot learning scheme for low-light image enhancement, which could train an exposure correction network named ExCNet without any prior information. It could enhance the low-light image by estimating the S-curve that best fits the image. Similarly, Guo et al. proposed a...
Existing low-light image enhancement techniques face challenges in achieving high visual quality and computational efficiency, as well as in effectively removing noise and adjusting illumination in extremely dark scenes. To address these problems, in thi
In [11], the light enhancement is creatively formulated as a task of image-specific curve estimation us- ing zero-shot learning. In [20, 47, 55], 3D lookup table and color histogram are utilized to preserve the color consis- tency. However, existing designs...
computer-visiondetectiontransformerimage-restorationlow-light-enhancelow-light-image-enhancementlow-light-vision UpdatedApr 28, 2025 Python Pytorch implementation of Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement deep-learningpytorchhdrzero-shot-learninglow-light-enhancelow-light-image-enh...
Image enhancement Low-light images Image processing Deep learning 1. Introduction Computer vision technology and deep learning are more and more widely used in many fields, such as medical image processing [1], automatic driving [2], face recognition [3], object detection [4]. However, due to...
International Journal of Computer Vision (2024) 132:4703–4723 https://doi.org/10.1007/s11263-024-02084-w Temporally Consistent Enhancement of Low-Light Videos via Spatial-Temporal Compatible Learning Lingyu Zhu1 · Wenhan Yang2 · Baoliang Chen1 · Hanwei Zhu1 · Xiandong Meng2 · Shiqi ...