Pixel-wise predictions are required in a wide variety of tasks such as image restoration, image segmentation, or disparity estimation. Common models involve several stages of data resampling, in which the resolution of feature maps is first reduced to aggregate information and then increased to generate a high-resolution output. Previous works have shown that resampling operations are subject to artifacts such as aliasing. During downsampling, aliases have been shown to compromise the prediction stability of image classifiers. During upsampling, they have been leveraged to detect generated content. Yet, the effect of aliases during upsampling has not yet been discussed w.r.t. the stability and robustness of pixel-wise predictions. While falling under the same term (aliasing), the challenges for correct upsampling in neural networks differ significantly from those during downsampling: when downsampling, some high frequencies can not be correctly represented and have to be removed to avoid aliases. However, when upsampling for pixel-wise predictions, we actually require the model to restore such high frequencies that can not be encoded in lower resolutions. The application of findings from signal processing is therefore a necessary but not a sufficient condition to achieve the desirable output. In contrast, we find that the availability of large spatial context during upsampling allows to provide stable, high-quality pixel-wise predictions, even when fully learning all filter weights.
DTC: A Deformable Transposed Convolution Module for Medical Image Segmentation
Department of Health Outcomes and Biomedical Informatics, University of Florida（健康结果与生物医学信息学系，佛罗里达大学）；Regenstrief Institue, Indiana University, IU Health（Regenstrief研究所，印第安纳大学，IU健康）
Escuela de Ingenieria y Ciencias, Tecnologico de Monterrey（蒙特雷理工学院工程与科学学院）；CRAN (UMR 7039), Université de Lorraine（洛林大学CRAN（UMR 7039））；CNRS（法国国家科学研究中心）
In this contribution we use an ensemble deep-learning method for combining the prediction of two individual one-stage detectors (i.e., YOLOv4 and Yolact) with the aim to detect artefacts in endoscopic images. This ensemble strategy enabled us to improve the robustness of the individual models without harming their real-time computation capabilities. We demonstrated the effectiveness of our approach by training and testing the two individual models and various ensemble configurations on the "Endoscopic Artifact Detection Challenge" dataset. Extensive experiments show the superiority, in terms o…
Deep convolutional neural networks (CNNs) have delivered superior performance in many computer vision tasks. In this paper, we propose a novel deep fully convolutional network model for accurate salient object detection. The key contribution of this work is to learn deep uncertain convolutional features (UCF), which encourage the robustness and accuracy of saliency detection. We achieve this via introducing a reformulated dropout (R-dropout) after specific convolutional layers to construct an uncertain ensemble of internal feature units. In addition, we propose an effective hybrid upsampling m…
Shanghai Jiao Tong University（上海交通大学）；Key Laboratory of System Control and Information Processing, Ministry of Education（教育部系统控制与信息处理重点实验室）；University of Shanghai for Science and Technology（上海理工大学）；University of Technology, Sydney（悉尼科技大学）；University of Delaware（特拉华大学）
Time-of-Flight (ToF) depth sensing camera is able to obtain depth maps at a high frame rate. However, its low resolution and sensitivity to the noise are always a concern. A popular solution is upsampling the obtained noisy low resolution depth map with the guidance of the companion high resolution color image. However, due to the constrains in the existing upsampling models, the high resolution depth map obtained in such way may suffer from either texture copy artifacts or blur of depth discontinuity. In this paper, a novel optimization framework is proposed with the brand new data term and s…
Dense prediction tasks typically employ encoder-decoder architectures, but the prevalent convolutions in the decoder are not image-adaptive and can lead to boundary artifacts. Different generalized convolution operations have been introduced to counteract this. We go beyond these by leveraging guidance data to redefine their inherent notion of proximity. Our proposed network layer builds on the permutohedral lattice, which performs sparse convolutions in a high-dimensional space allowing for powerful non-local operations despite small filters. Multiple features with different characteristics s…
Depth completion involves estimating a dense depth image from sparse depth measurements, often guided by a color image. While linear upsampling is straight forward, it results in artifacts including depth pixels being interpolated in empty space across discontinuities between objects. Current methods use deep networks to upsample and "complete" the missing depth pixels. Nevertheless, depth smearing between objects remains a challenge. We propose a new representation for depth called Depth Coefficients (DC) to address this problem. It enables convolutions to more easily avoid inter-object depth…
We propose a novel method of efficient upsampling of a single natural image. Current methods for image upsampling tend to produce high-resolution images with either blurry salient edges, or loss of fine textural detail, or spurious noise artifacts. In our method, we mitigate these effects by modeling the input image as a sum of edge and detail layers, operating upon these layers separately, and merging the upscaled results in an automatic fashion. We formulate the upsampled output image as the solution to a non-convex energy minimization problem, and propose an algorithm to obtain a tractable …
Liquid Fusion of Heterogeneous Representations Towards General Salient Object Detection
School of Computer Science and Artificial Intelligence, Changzhou University（常州大学计算机科学与人工智能学院）；College of Computer Science and Artificial Intelligence, Fudan University（复旦大学计算机科学与人工智能学院）；Faculty of Data Science, City University of Macau（澳门城市大学数据科学学院）
Guilin University of Electronic Technology（桂林电子科技大学）
Semantic segmentation is applied extensively in autonomous driving and intelligent transportation with methods that highly demand spatial and semantic information. Here, an STDC-MA network is proposed to meet these demands. First, the STDC-Seg structure is employed in STDC-MA to ensure a lightweight and efficient structure. Subsequently, the feature alignment module (FAM) is applied to understand the offset between high-level and low-level features, solving the problem of pixel offset related to upsampling on the high-level feature map. Our approach implements the effective fusion between high…
The Graduate Center of the City University of New York（纽约城市大学研究生中心）；Borough of Manhattan Community College - CUNY（纽约城市大学曼哈顿社区学院）；The City College of New York - CUNY（纽约城市大学纽约城市学院）
Gatherings of thousands to millions of people frequently occur for an enormous variety of events, and automated counting of these high-density crowds is useful for safety, management, and measuring significance of an event. In this work, we show that the regularly accepted labeling scheme of crowd density maps for training deep neural networks is less effective than our alternative inverse k-nearest neighbor (i$k$NN) maps, even when used directly in existing state-of-the-art network structures. We also provide a new network architecture MUD-i$k$NN, which uses multi-scale upsampling via transpo…
Autonomous Vision Group（自主视觉小组）；University of Tübingen（图宾根大学）；MPI for Intelligent Systems（马克斯·普朗克智能系统研究所）
The key objective of Generative Adversarial Networks (GANs) is to generate new data with the same statistics as the provided training data. However, multiple recent works show that state-of-the-art architectures yet struggle to achieve this goal. In particular, they report an elevated amount of high frequencies in the spectral statistics which makes it straightforward to distinguish real and generated images. Explanations for this phenomenon are controversial: While most works attribute the artifacts to the generator, other works point to the discriminator. We take a sober look at those explan…
MBPU: A Plug-and-Play State Space Model for Point Cloud Upsamping with Fast Point Rendering
School of Computer Science, Fudan University（复旦大学计算机科学学院）；Academy for Engineering and Technology, Fudan University（复旦大学工程技术研究院）；Department of Information Engineering, The Chinese University of Hong Kong（香港中文大学信息工程系）；School of Mechanical Engineering, Tongji University（同济大学机械工程学院）
This paper explores an efficient solution for Space-time Super-Resolution, aiming to generate High-resolution Slow-motion videos from Low Resolution and Low Frame rate videos. A simplistic solution is the sequential running of Video Super Resolution and Video Frame interpolation models. However, this type of solutions are memory inefficient, have high inference time, and could not make the proper use of space-time relation property. To this extent, we first interpolate in LR space using quadratic modeling. Input LR frames are super-resolved using a state-of-the-art Video Super-Resolution metho…
Dalian University of Technology（大连理工大学）；Adobe Research（Adobe研究院）
Existing image inpainting methods often produce artifacts when dealing with large holes in real applications. To address this challenge, we propose an iterative inpainting method with a feedback mechanism. Specifically, we introduce a deep generative model which not only outputs an inpainting result but also a corresponding confidence map. Using this map as feedback, it progressively fills the hole by trusting only high-confidence pixels inside the hole at each iteration and focuses on the remaining pixels in the next iteration. As it reuses partial predictions from the previous iterations as …
In this paper, we propose a novel convolutional neural network (CNN) that never causes checkerboard artifacts, for image enhancement. In research fields of image-to-image translation problems, it is well-known that images generated by usual CNNs are distorted by checkerboard artifacts which mainly caused in forward-propagation of upsampling layers. However, checkerboard artifacts in image enhancement have never been discussed. In this paper, we point out that applying U-Net based CNNs to image enhancement causes checkerboard artifacts. In contrast, the proposed network that contains fixed conv…
Single-image super-resolution (SR) with fixed and discrete scale factors has achieved great progress due to the development of deep learning technology. However, the continuous-scale SR, which aims to use a single model to process arbitrary (integer or non-integer) scale factors, is still a challenging task. The existing SR models generally adopt static convolution to extract features, and thus unable to effectively perceive the change of scale factor, resulting in limited generalization performance on multi-scale SR tasks. Moreover, the existing continuous-scale upsampling modules do not make…
Improving Quality Control of Whole Slide Images by Explicit Artifact Augmentation
AGH University of Krakow（克拉科夫AGH科技大学）；University of Applied Sciences Western Switzerland (HES-SO)（瑞士西部应用科学大学（HES-SO））；Radboud University Medical Center（拉德堡德大学医学中心）；University of Padova（帕多瓦大学）；University of Geneva（日内瓦大学）
University College London（伦敦大学学院）；Google Inc.（谷歌公司）
Many machine vision applications, such as semantic segmentation and depth prediction, require predictions for every pixel of the input image. Models for such problems usually consist of encoders which decrease spatial resolution while learning a high-dimensional representation, followed by decoders who recover the original input resolution and result in low-dimensional predictions. While encoders have been studied rigorously, relatively few studies address the decoder side. This paper presents an extensive comparison of a variety of decoders for a variety of pixel-wise tasks ranging from class…
Imaging Biomarkers for Neurodegenerative Diseases from Detailed Segmentation of Medial Temporal Lobe Subregions on in vivo Brain MRI Using Upsampling Strategy Guided by High-resolution ex vivo MRI
University of Pennsylvania（宾夕法尼亚大学）；Siemens Healthineers（西门子医疗）；Lund University（隆德大学）；University of Castilla La Mancha（卡斯蒂利亚-拉曼恰大学）
Deep neural networks can generate images that are astonishingly realistic, so much so that it is often hard for humans to distinguish them from actual photos. These achievements have been largely made possible by Generative Adversarial Networks (GANs). While deep fake images have been thoroughly investigated in the image domain - a classical approach from the area of image forensics - an analysis in the frequency domain has been missing so far. In this paper, we address this shortcoming and our results reveal that in frequency space, GAN-generated images exhibit severe artifacts that can be ea…
Tiling artifacts and trade-offs of feature normalization in the segmentation of large biological images
European Molecular Biology Laboratory（欧洲分子生物学实验室）；Institute of Computer Science, University of Göttingen（哥廷根大学计算机科学学院）；IRCCS Humanitas Research Hospital（Humanitas 医院（IRCCS））
Different from traditional hyperspectral super-resolution approaches that focus on improving the spatial resolution, spectral super-resolution aims at producing a high-resolution hyperspectral image from the RGB observation with super-resolution in spectral domain. However, it is challenging to accurately reconstruct a high-dimensional continuous spectrum from three discrete intensity values at each pixel, since too much information is lost during the procedure where the latent hyperspectral image is downsampled (e.g., with x10 scaling factor) in spectral domain to produce an RGB observation. …
Frequency-Aware Deepfake Detection: Improving Generalizability through Frequency Space Learning
Institute of Information Science, Beijing Jiaotong University（北京交通大学信息科学研究所）；School of Information Science and Engineering, Yanshan University（燕山大学信息科学与工程学院）；Center for Frontier AI Research, IHPC, A*STAR, Singapore（新加坡科技研究局前沿人工智能研究中心）
Ultrasound imaging has been improving, but continues to suffer from inherent artifacts that are challenging to model, such as attenuation, shadowing, diffraction, speckle, etc. These artifacts can potentially confuse image analysis algorithms unless an attempt is made to assess the certainty of individual pixel values. Our novel confidence algorithms analyze pixel values using a directed acyclic graph based on acoustic physical properties of ultrasound imaging. We demonstrate unique capabilities of our approach and compare it against previous confidence-measurement algorithms for shadow-detect…
A Steel Surface Defect Detection Method Based on Lightweight Convolution Optimization
Hainan Tropical Ocean University（海南热带海洋学院）；University of Sanya（三亚学院）；Universiti Malaysia Perlis（马来西亚玻璃市大学）；Thalesgroup（泰雷兹集团）；Pukyong National University（釜庆国立大学）
Ultrasound image quality has continually been improving. However, when needles or other metallic objects are operating inside the tissue, the resulting reverberation artifacts can severely corrupt the surrounding image quality. Such effects are challenging for existing computer vision algorithms for medical image analysis. Needle reverberation artifacts can be hard to identify at times and affect various pixel values to different degrees. The boundaries of such artifacts are ambiguous, leading to disagreement among human experts labeling the artifacts. We propose a weakly- and semi-supervised,…
Improving Text-Instance Alignment Of Foreground Conditioned Out-Painting Via Customized Concept Embedding
College of Electronic and Information Engineering, Tongji University, Shanghai, China（电子信息工程学院，同济大学，上海，中国）；State Key Laboratory of Autonomous Intelligent Unmanned Systems, Frontiers Science Center for Intelligent Autonomous Systems, Ministry of Education, Shanghai, China（自主智能无人系统国家重点实验室，智能自主系统前沿科学中心，教育部，上海，中国）
