Abstract:In Collaborative Intelligence, a deep neural network (DNN) is partitioned and deployed at the edge and the cloud for bandwidth saving and system optimization. When a model input is an image, it has been confirmed that the intermediate feature map, the output from the edge, can be smaller than the input data size. However, its effectiveness has not been reported when the input is a video. In this study, we propose a method to compress the feature map of surveillance videos by applying inter-feature-map differential coding (IFMDC). IFMDC shows a compression ratio comparable to, or better than, HEVC to the input video in the case of small accuracy reduction. Our method is especially effective for videos that are sensitive to image quality degradation when HEVC is applied
Abstract:Image Coding for Machines (ICM) is becoming more important as research in computer vision progresses. ICM is a vital research field that pursues the use of images for image recognition models, facilitating efficient image transmission and storage. The demand for recognition models is growing rapidly among the general public, and their performance continues to improve. To meet these needs, exchanging image data between consumer devices and cloud AI using ICM technology could be one possible solution. In ICM, various image compression methods have adopted Learned Image Compression (LIC). LIC includes an entropy model for estimating the bitrate of latent features, and the design of this model significantly affects its performance. Typically, LIC methods assume that the distribution of latent features follows a normal distribution. This assumption is effective for compressing images intended for human vision. However, employing an entropy model based on normal distribution is inefficient in ICM due to the limitation of image parts that require precise decoding. To address this, we propose Delta-ICM, which uses a probability distribution based on a delta function. Assuming the delta distribution as a distribution of latent features reduces the entropy of image portions unnecessary for machines. We compress the remaining portions using an entropy model based on normal distribution, similar to existing methods. Delta-ICM selects between the entropy model based on the delta distribution and the one based on the normal distribution for each latent feature. Our method outperforms existing ICM methods in image compression performance aimed at machines.
Abstract:Implicit neural representations (INRs) embed various signals into networks. They have gained attention in recent years because of their versatility in handling diverse signal types. For videos, INRs achieve video compression by embedding video signals into networks and compressing them. Conventional methods use an index that expresses the time of the frame or the features extracted from the frame as inputs to the network. The latter method provides greater expressive capability as the input is specific to each video. However, the features extracted from frames often contain redundancy, which contradicts the purpose of video compression. Moreover, since frame time information is not explicitly provided to the network, learning the relationships between frames is challenging. To address these issues, we aim to reduce feature redundancy by extracting features based on the high-frequency components of the frames. In addition, we use feature differences between adjacent frames in order for the network to learn frame relationships smoothly. We propose a video representation method that uses the high-frequency components of frames and the differences in features between adjacent frames. The experimental results show that our method outperforms the existing HNeRV method in 90 percent of the videos.
Abstract:Scalable image coding for both humans and machines is a technique that has gained a lot of attention recently. This technology enables the hierarchical decoding of images for human vision and image recognition models. It is a highly effective method when images need to serve both purposes. However, no research has yet incorporated the post-processing commonly used in popular image compression schemes into scalable image coding method for humans and machines. In this paper, we propose a method to enhance the quality of decoded images for humans by integrating post-processing into scalable coding scheme. Experimental results show that the post-processing improves compression performance. Furthermore, the effectiveness of the proposed method is validated through comparisons with traditional methods.
Abstract:Image Coding for Machines (ICM) is an image compression technique for image recognition. This technique is essential due to the growing demand for image recognition AI. In this paper, we propose a method for ICM that focuses on encoding and decoding only the edge information of object parts in an image, which we call SA-ICM. This is an Learned Image Compression (LIC) model trained using edge information created by Segment Anything. Our method can be used for image recognition models with various tasks. SA-ICM is also robust to changes in input data, making it effective for a variety of use cases. Additionally, our method provides benefits from a privacy point of view, as it removes human facial information on the encoder's side, thus protecting one's privacy. Furthermore, this LIC model training method can be used to train Neural Representations for Videos (NeRV), which is a video compression model. By training NeRV using edge information created by Segment Anything, it is possible to create a NeRV that is effective for image recognition (SA-NeRV). Experimental results confirm the advantages of SA-ICM, presenting the best performance in image compression for image recognition. We also show that SA-NeRV is superior to ordinary NeRV in video compression for machines.
Abstract:Image coding for machines (ICM) aims to compress images for machine analysis using recognition models rather than human vision. Hence, in ICM, it is important for the encoder to recognize and compress the information necessary for the machine recognition task. There are two main approaches in learned ICM; optimization of the compression model based on task loss, and Region of Interest (ROI) based bit allocation. These approaches provide the encoder with the recognition capability. However, optimization with task loss becomes difficult when the recognition model is deep, and ROI-based methods often involve extra overhead during evaluation. In this study, we propose a novel training method for learned ICM models that applies auxiliary loss to the encoder to improve its recognition capability and rate-distortion performance. Our method achieves Bjontegaard Delta rate improvements of 27.7% and 20.3% in object detection and semantic segmentation tasks, compared to the conventional training method.
Abstract:The digitalization of society is rapidly developing toward the realization of the digital twin and metaverse. In particular, point clouds are attracting attention as a media format for 3D space. Point cloud data is contaminated with noise and outliers due to measurement errors. Therefore, denoising and outlier detection are necessary for point cloud processing. Among them, PointCleanNet is an effective method for point cloud denoising and outlier detection. However, it does not consider the local geometric structure of the patch. We solve this problem by applying two types of graph convolutional layer designed based on the Dynamic Graph CNN. Experimental results show that the proposed methods outperform the conventional method in AUPR, which indicates outlier detection accuracy, and Chamfer Distance, which indicates denoising accuracy.
Abstract:Compression technology is essential for efficient image transmission and storage. With the rapid advances in deep learning, images are beginning to be used for image recognition as well as for human vision. For this reason, research has been conducted on image coding for image recognition, and this field is called Image Coding for Machines (ICM). There are two main approaches in ICM: the ROI-based approach and the task-loss-based approach. The former approach has the problem of requiring an ROI-map as input in addition to the input image. The latter approach has the problems of difficulty in learning the task-loss, and lack of robustness because the specific image recognition model is used to compute the loss function. To solve these problems, we propose an image compression model that learns object regions. Our model does not require additional information as input, such as an ROI-map, and does not use task-loss. Therefore, it is possible to compress images for various image recognition models. In the experiments, we demonstrate the versatility of the proposed method by using three different image recognition models and three different datasets. In addition, we verify the effectiveness of our model by comparing it with previous methods.
Abstract:Generally, Deep Neural Networks (DNNs) are expected to have high performance when their model size is large. However, large models failed to produce high-quality results commensurate with their scale in music Super-Resolution (SR). We attribute this to that DNNs cannot learn information commensurate with their size from standard mean square error losses. To unleash the potential of large DNN models in music SR, we propose BigWavGAN, which incorporates Demucs, a large-scale wave-to-wave model, with State-Of-The-Art (SOTA) discriminators and adversarial training strategies. Our discriminator consists of Multi-Scale Discriminator (MSD) and Multi-Resolution Discriminator (MRD). During inference, since only the generator is utilized, there are no additional parameters or computational resources required compared to the baseline model Demucs. Objective evaluation affirms the effectiveness of BigWavGAN in music SR. Subjective evaluations indicate that BigWavGAN can generate music with significantly high perceptual quality over the baseline model. Notably, BigWavGAN surpasses the SOTA music SR model in both simulated and real-world scenarios. Moreover, BigWavGAN represents its superior generalization ability to address out-of-distribution data. The conducted ablation study reveals the importance of our discriminators and training strategies. Samples are available on the demo page.
Abstract:Audio Super-Resolution (SR) is an important topic in the field of audio processing. Many models are designed in time domain due to the advantage of waveform processing, such as being able to avoid the phase problem. However, in prior works it is shown that Time-Domain Convolutional Neural Network (TD-CNN) approaches tend to produce annoying artifacts in their output. In order to confirm the source of the artifact, we conduct an AB listening test and found phase to be the cause. We further propose Time-Domain Phase Repair (TD-PR) to improve TD-CNNs' performance by repairing the phase of the TD-CNNs' output. In this paper, we focus on the music SR task, which is challenging due to the wide frequency response and dynamic range of music. Our proposed method can handle various narrow-bandwidth from 2.5kHz to 4kHz with a target bandwidth of 8kHz. We conduct both objective and subjective evaluation to assess the proposed method. The objective evaluation result indicates the proposed method achieves the SR task effectively. Moreover, the proposed TD-PR obtains the much higher mean opinion scores than all TD-CNN baselines, which indicates that the proposed TD-PR significantly improves perceptual quality. Samples are available on the demo page.