We have developed and tested in numerical experiments a universal approach to searching objects of a given type in captured video images (for example, people's faces, vehicles, special characters, numbers and letters, etc.). The novelty and versatility of this approach consists in a unique combination of the well-known methods ranging from creating detectors to making decisions independent of the type of recognition objects. The efficiencies of various types of basic features used for image coding, including the Haar features, the LBP features, and the modified Census transformation are compared. A combination of the modified methods is used for constructing 11 types of detectors of the number of railway carriages and for recognizing digits from zero to nine. The efficiency of the constructed detectors is studied.
With substantial public concerns on potential cancer risks and health hazards caused by the accumulated radiation exposure in medical imaging, reducing radiation dose in X-ray based medical imaging such as Computed Tomography Perfusion (CTP) has raised significant research interests. In this paper, we embrace the deep Convolutional Neural Networks (CNN) based approaches and introduce Smoothed Dense-Convolution Neural Network (SDCNet) to recover high-dose quality CTP images from low-dose ones. SDCNet is composed of sub-network blocks cascaded by skip-connections to infer the noise (differentials) from paired low/high-dose CT scans. SDCNet can effectively remove the noise in real low-dose CT scans and enhance the quality of medical images. We evaluate the proposed architecture on thousands of CT perfusion frames for both reconstructed image denoising and perfusion map quantification including cerebral blood flow (CBF) and cerebral blood volume (CBV). SDCNet achieves high performance in both visual and quantitative results with promising computational efficiency, comparing favorably with state-of-the-art approaches. \textit{The code is available at \url{https://github.com/cswin/RC-Nets}}.
In this work, we introduce Panoptic-DeepLab, a simple, strong, and fast system for panoptic segmentation, aiming to establish a solid baseline for bottom-up methods that can achieve comparable performance of two-stage methods while yielding fast inference speed. In particular, PanopticDeepLab adopts the dual-ASPP and dual-decoder structures specific to semantic, and instance segmentation, respectively. The semantic segmentation branch is the same as the typical design of any semantic segmentation model (e.g., DeepLab), while the instance segmentation branch is class-agnostic, involving a simple instance center regression. As a result, our single Panoptic-DeepLab simultaneously ranks first at all three Cityscapes benchmarks, setting the new state-of-art of 84.2% mIoU, 39.0% AP, and 65.5% PQ on test set. Additionally, equipped with MobileNetV3, Panoptic-DeepLab runs nearly in real-time with a single 1025 x 2049 image (15.8 frames per second), while achieving a competitive performance on Cityscapes (54.1 PQ% on test set). On Mapillary Vistas test set, our ensemble of six models attains 42.7% PQ, outperforming the challenge winner in 2018 by a healthy margin of 1.5%. Finally, our Panoptic-DeepLab also performs on par with several topdown approaches on the challenging COCO dataset. For the first time, we demonstrate a bottom-up approach could deliver state-of-the-art results on panoptic segmentation.
In recent years, long short-term memory neural networks (LSTMs) have been applied quite successfully to problems in handwritten text recognition. However, their strength is more located in handling sequences of variable length than in handling geometric variability of the image patterns. Furthermore, the best results for LSTMs are often based on large-scale training of an ensemble of network instances. In this paper, an end-to-end convolutional LSTM Neural Network is used to handle both geometric variation and sequence variability. We show that high performances can be reached on a common benchmark set by using proper data augmentation for just five such networks using a proper coding scheme and a proper voting scheme. The networks have similar architectures (Convolutional Neural Network (CNN): five layers, bidirectional LSTM (BiLSTM): three layers followed by a connectionist temporal classification (CTC) processing step). The approach assumes differently-scaled input images and different feature map sizes. Two datasets are used for evaluation of the performance of our algorithm: A standard benchmark RIMES dataset (French), and a historical handwritten dataset KdK (Dutch). Final performance obtained for the word-recognition test of RIMES was 96.6%, a clear improvement over other state-of-the-art approaches. On the KdK dataset, our approach also shows good results. The proposed approach is deployed in the Monk search engine for historical-handwriting collections.
Recent works attempt to improve scene parsing performance by exploring different levels of contexts, and typically train a well-designed convolutional network to exploit useful contexts across all pixels equally. However, in this paper, we find that the context demands are varying from different pixels or regions in each image. Based on this observation, we propose an Adaptive Context Network (ACNet) to capture the pixel-aware contexts by a competitive fusion of global context and local context according to different per-pixel demands. Specifically, when given a pixel, the global context demand is measured by the similarity between the global feature and its local feature, whose reverse value can be used to measure the local context demand. We model the two demand measurements by the proposed global context module and local context module, respectively, to generate adaptive contextual features. Furthermore, we import multiple such modules to build several adaptive context blocks in different levels of network to obtain a coarse-to-fine result. Finally, comprehensive experimental evaluations demonstrate the effectiveness of the proposed ACNet, and new state-of-the-arts performances are achieved on all four public datasets, i.e. Cityscapes, ADE20K, PASCAL Context, and COCO Stuff.
Today, image classification is a common way for systems to process visual content. Although neural network approaches to classification have seen great progress in reducing error rates, it is not clear what this means for a cognitive system that needs to make sense of the multiple and competing predictions from its own classifiers. As a step to address this, we present a novel framework that uses meta-reasoning and meta-operations to unify predictions into abstractions, properties, or relationships. Using the framework on images from ImageNet, we demonstrate systems that unify 41% to 46% of predictions in general and unify 67% to 75% of predictions when the systems can explain their conceptual differences. We also demonstrate a system in "the wild" by feeding live video images through it and show it unifying 51% of predictions in general and 69% of predictions when their differences can be explained conceptually by the system. In a survey given to 24 participants, we found that 87% of the unified predictions describe their corresponding images.
This paper addresses the single-image compressive sensing (CS) and reconstruction problem. We propose a scalable Laplacian pyramid reconstructive adversarial network (LAPRAN) that enables high-fidelity, flexible and fast CS images reconstruction. LAPRAN progressively reconstructs an image following the concept of Laplacian pyramid through multiple stages of reconstructive adversarial networks (RANs). At each pyramid level, CS measurements are fused with a contextual latent vector to generate a high-frequency image residual. Consequently, LAPRAN can produce hierarchies of reconstructed images and each with an incremental resolution and improved quality. The scalable pyramid structure of LAPRAN enables high-fidelity CS reconstruction with a flexible resolution that is adaptive to a wide range of compression ratios (CRs), which is infeasible with existing methods. Experimental results on multiple public datasets show that LAPRAN offers an average 7.47dB and 5.98dB PSNR, and an average 57.93% and 33.20% SSIM improvement compared to model-based and data-driven baselines, respectively.
In this work we propose to improve text recognition from a new perspective by separating text content from complex backgrounds. We exploit the generative adversarial networks (GANs) for removing backgrounds while retaining the text content. As vanilla GANs are not sufficiently robust to generate sequence-like characters in natural images, we propose an adversarial learning framework for the generation and recognition of multiple characters in an image. The proposed framework consists of an attention-based recognizer and a generative adversarial architecture. Furthermore, to tackle the lack of paired training samples, we design an interactive joint training scheme, which shares attention masks from the recognizer to the discriminator, and enables the discriminator to extract the features of every character for further adversarial training. Benefiting from the character-level adversarial training, our framework requires only unpaired simple data for style supervision. Every target style sample containing only one randomly chosen character can be simply synthesized online during the training. This is significant as the training does not require costly paired samples or character-level annotations. Thus, only the input images and corresponding text labels are needed. In addition to the style transfer of the backgrounds, we refine character patterns to ease the recognition task. A feedback mechanism is proposed to bridge the gap between the discriminator and the recognizer. Therefore, the discriminator can guide the generator according to the confusion of the recognizer. The generated patterns are thus clearer for recognition. Experiments on various benchmarks, including both regular and irregular text, demonstrate that our method significantly reduces the difficulty of recognition. Our framework can be integrated with recent recognition methods to achieve new state-of-the-art performance.
The premise of our work is deceptively familiar: A black box $f(\cdot)$ has altered an image $\mathbf{x} \rightarrow f(\mathbf{x})$. Recover the image $\mathbf{x}$. This black box might be any number of simple or complicated things: a linear or non-linear filter, some app on your phone, etc. The latter is a good canonical example for the problem we address: Given only "the app" and an image produced by the app, find the image that was fed to the app. You can run the given image (or any other image) through the app as many times as you like, but you can not look inside the (code for the) app to see how it works. At first blush, the problem sounds a lot like a standard inverse problem, but it is not in the following sense: While we have access to the black box $f(\cdot)$ and can run any image through it and observe the output, we do not know how the block box alters the image. Therefore we have no explicit form or model of $f(\cdot)$. Nor are we necessarily interested in the internal workings of the black box. We are simply happy to reverse its effect on a particular image, to whatever extent possible. This is what we call the "rendition" (rather than restoration) problem, as it does not fit the mold of an inverse problem (blind or otherwise). We describe general conditions under which rendition is possible, and provide a remarkably simple algorithm that works for both contractive and expansive black box operators. The principal and novel take-away message from our work is this surprising fact: One simple algorithm can reliably undo a wide class of (not too violent) image distortions. A higher quality pdf of this paper is available at http://www.milanfar.org
It is known that deep neural networks (DNNs) could be vulnerable to adversarial attacks. The so-called physical adversarial examples deceive DNN-based decision makers by attaching adversarial patches to real objects. However, most of the existing works on physical adversarial attacks focus on static objects such as glass frame, stop sign and image attached to a cardboard. In this work, we proposed adversarial T-shirt, a robust physical adversarial example for evading person detectors even if it suffers from deformation due toa moving person's pose change. To the best of our knowledge, the effect of deformation is first modeled for designing physical adversarial examples with respect to non-rigid objects such as T-shirts. We show that the proposed method achieves 79% and 63% attack success rates in digital and physical worlds respectively against YOLOv2. In contrast, the state-of-the-art physical attack method to fool a person detector only achieves 27% attack success rate. Furthermore, by leveraging min-max optimization, we extend our method to the ensemble attack setting against object detectors YOLOv2 and Faster R-CNN simultaneously.