One of the main difficulties of scaling current localization systems to large environments is the on-board storage required for the maps. In this paper we propose to learn to compress the map representation such that it is optimal for the localization task. As a consequence, higher compression rates can be achieved without loss of localization accuracy when compared to standard coding schemes that optimize for reconstruction, thus ignoring the end task. Our experiments show that it is possible to learn a task-specific compression which reduces storage requirements by two orders of magnitude over general-purpose codecs such as WebP without sacrificing performance.
We address the challenge of applying existing convolutional neural network (CNN) architectures to compressed images. Existing CNN architectures represent images as a matrix of pixel intensities with a specified dimension; this desired dimension is achieved by downgrading or cropping. Downgrading and cropping are attractive in that the result is also an image; however, an algorithm producing an alternative "compressed" representation could yield better classification performance. This compression algorithm need not be reversible, but must be compatible with the CNN's operations. This problem is…
Xian Jiaotong University（西安交通大学）；The Chinese University of Hong Kong（香港中文大学）
The 3-D total variation (3DTV) is a powerful regularization term, which encodes the local smoothness prior structure underlying a hyper-spectral image (HSI), for general HSI processing tasks. This term is calculated by assuming identical and independent sparsity structures on all bands of gradient maps calculated along spatial and spectral HSI modes. This, however, is always largely deviated from the real cases, where the gradient maps are generally with different while correlated sparsity structures across all their bands. Such deviation tends to hamper the performance of the related method b…
Convolutional Neural Networks (CNNs) are known for requiring extensive computational resources, and quantization is among the best and most common methods for compressing them. While aggressive quantization (i.e., less than 4-bits) performs well for classification, it may cause severe performance degradation in image-to-image tasks such as semantic segmentation and depth estimation. In this paper, we propose Wavelet Compressed Convolution (WCC) -- a novel approach for high-resolution activation maps compression integrated with point-wise convolutions, which are the main computational cost of m…
Zhejiang University（浙江大学）；Arizona State University（亚利桑那州立大学）
Incorporating semantic information into the codecs during image compression can significantly reduce the repetitive computation of fundamental semantic analysis (such as object recognition) in client-side applications. The same practice also enable the compressed code to carry the image semantic information during storage and transmission. In this paper, we propose a concept called Deep Semantic Image Compression (DeepSIC) and put forward two novel architectures that aim to reconstruct the compressed image and generate corresponding semantic representations at the same time. The first architec…
Mangalore University（芒格洛尔大学）；KVG College of Engineering（KVG工程学院）
In this paper, we propose a gradient difference based approach to text localization in videos and scene images. The input video frame/ image is first compressed using multilevel 2-D wavelet transform. The edge information of the reconstructed image is found which is further used for finding the maximum gradient difference between the pixels and then the boundaries of the detected text blocks are computed using zero crossing technique. We perform logical AND operation of the text blocks obtained by gradient difference and the zero crossing technique followed by connected component analysis to e…
Motivated by recent work on deep neural network (DNN)-based image compression methods showing potential improvements in image quality, savings in storage, and bandwidth reduction, we propose to perform image understanding tasks such as classification and segmentation directly on the compressed representations produced by these compression methods. Since the encoders and decoders in DNN-based compression methods are neural networks with feature-maps as internal representations of the images, we directly integrate these with architectures for image understanding. This bypasses decoding of the co…
Uber Advanced Technologies Group（优步先进技术集团）；MIT（麻省理工学院）；University of Toronto（多伦多大学）
In this paper we propose a novel semantic localization algorithm that exploits multiple sensors and has precision on the order of a few centimeters. Our approach does not require detailed knowledge about the appearance of the world, and our maps require orders of magnitude less storage than maps utilized by traditional geometry- and LiDAR intensity-based localizers. This is important as self-driving cars need to operate in large environments. Towards this goal, we formulate the problem in a Bayesian filtering framework, and exploit lanes, traffic signs, as well as vehicle dynamics to localize …
MILA（蒙特利尔学习算法研究所）；Université de Montréal（蒙特利尔大学）；Imagia Inc.（Imagia公司）；School of Computing Science, Simon Fraser University（西蒙弗雷泽大学计算科学学院）
The scarcity of richly annotated medical images is limiting supervised deep learning based solutions to medical image analysis tasks, such as localizing discriminatory radiomic disease signatures. Therefore, it is desirable to leverage unsupervised and weakly supervised models. Most recent weakly supervised localization methods apply attention maps or region proposals in a multiple instance learning formulation. While attention maps can be noisy, leading to erroneously highlighted regions, it is not simple to decide on an optimal window/bag size for multiple instance learning approaches. In th…
Hierarchical Frequency-based Upsampling and Refining for Compressed Video Quality Enhancement
Hangzhou Dianzi University（杭州电子科技大学）；State Key Laboratory of Media Convergence Production Technology and Systems（媒体融合生产技术与系统国家重点实验室）；Xinhua Zhiyun Technology Co., Ltd.（新华智云科技有限公司）
Institute of Medical Technology, Peking University（北京大学医学技术研究院）；Peking University（北京大学）；Institute of Biomedical Engineering, Peking University Shenzhen Graduate School（北京大学深圳研究生院生物医学工程研究所）
Classification activation map (CAM), utilizing the classification structure to generate pixel-wise localization maps, is a crucial mechanism for weakly supervised object localization (WSOL). However, CAM directly uses the classifier trained on image-level features to locate objects, making it prefers to discern global discriminative factors rather than regional object cues. Thus only the discriminative locations are activated when feeding pixel-level features into this classifier. To solve this issue, this paper elaborates a plug-and-play mechanism called BagCAMs to better project a well-train…
Enhancing the Quality of 3D Lunar Maps Using JAXA's Kaguya Imagery
Jet Propulsion Laboratory, California Institute of Technology（喷气推进实验室，加州理工学院）；Department of Informatics, University of Oslo（信息学院，奥斯陆大学）；LunaSol Space LLC（LunaSol空间有限公司）；Department of Information Science and Electrical Engineering , Kyushu University（信息科学与电气工程学院，九州大学）
The deep learning revolution brought us an extensive array of neural network architectures that achieve state-of-the-art performance in a wide variety of Computer Vision tasks including among others, classification, detection and segmentation. In parallel, we have also been observing an unprecedented demand in computational and memory requirements, rendering the efficient use of neural networks in low-powered devices virtually unattainable. Towards this end, we propose a three-stage compression and acceleration pipeline that sparsifies, quantizes and entropy encodes activation maps of Convolut…
KU Leuven（鲁汶大学）；Ford（福特公司）；Ford Greenfield Labs（福特格林菲尔德实验室）
We describe a Deep-Geometric Localizer that is able to estimate the full 6 Degree of Freedom (DoF) global pose of the camera from a single image in a previously mapped environment. Our map is a topo-metric one, with discrete topological nodes whose 6 DoF poses are known. Each topo-node in our map also comprises of a set of points, whose 2D features and 3D locations are stored as part of the mapping process. For the mapping phase, we utilise a stereo camera and a regular stereo visual SLAM pipeline. During the localization phase, we take a single camera image, localize it to a topological node …
