The JPEG is one of the most widely used lossy image-compression standards, whose compression performance depends largely on a quantization table. In this work, we utilize a Convolutional Neural Network (CNN) to generate an image-adaptive quantization table in a standard-compliant way. We first build an image set containing more than 10,000 images and generate their optimal quantization tables through a classical genetic algorithm, and then propose a method that can efficiently extract and fuse the frequency and spatial domain information of each image to train a regression network to directly generate adaptive quantization tables. In addition, we extract several representative quantization tables from the dataset and train a classification network to indicate the optimal one for each image, which further improves compression performance and computational efficiency. Tests on diverse images show that the proposed method clearly outperforms the state-of-the-art method. Compared with the standard table at the compression rate of 1.0 bpp, the regression and classification network provide average Peak Signal-to-Noise Ratio (PSNR) gains of nearly 1.2 and 1.4 dB. For the experiment under Structural Similarity Index Measurement (SSIM), the improvements are 0.4% and 0.54%, respectively. The proposed method also has competitive computational efficiency, as the regression and classification network only take 15 and 6.25 milliseconds, respectively, to process a $768 \times 512$ image on a single CPU core at 3.20 GHz.

中文翻译：

---

QNet：基于卷积神经网络的自适应量化表生成器

JPEG是使用最广泛的有损图像压缩标准之一，其压缩性能很大程度上取决于量化表。在这项工作中，我们利用卷积神经网络（CNN）以符合标准的方式生成图像自适应量化表。我们首先构建包含超过10,000张图像的图像集，并通过经典遗传算法生成它们的最佳量化表，然后提出一种可以有效提取和融合每幅图像的频率和空间域信息以训练回归网络以直接进行建模的方法生成自适应量化表。此外，我们从数据集中提取了几个代表性的量化表，并训练了一个分类网络，以指示每个图像的最佳量化表，从而进一步提高了压缩性能和计算效率。对各种图像的测试表明，所提出的方法明显优于最新方法。与标准表在1.0 bpp的压缩率相比，回归和分类网络提供的平均峰值信噪比（PSNR）增益接近1.2和1.4 dB。对于结构相似指数测量（SSIM）下的实验，改进分别为0.4％和0.54％。所提出的方法还具有竞争性的计算效率，因为回归和分类网络分别只需要15和6.25毫秒即可处理 0 bpp时，回归和分类网络可提供近1.2和1.4 dB的平均峰值信噪比（PSNR）增益。对于结构相似指数测量（SSIM）下的实验，改进分别为0.4％和0.54％。所提出的方法还具有竞争性的计算效率，因为回归和分类网络分别只需要15和6.25毫秒即可处理 0 bpp时，回归和分类网络可提供近1.2和1.4 dB的平均峰值信噪比（PSNR）增益。对于结构相似指数测量（SSIM）下的实验，改进分别为0.4％和0.54％。所提出的方法还具有竞争性的计算效率，因为回归和分类网络分别只需要15和6.25毫秒即可处理 768美元/ 512美元 在3.20 GHz的单个CPU内核上显示图像。