Image quality assessment (IQA) is traditionally classified into full-reference (FR) IQA and no-reference (NR) IQA according to whether the original image is required. Although NR-IQA is widely used in practical applications, room for improvement still remains because of the lack of the reference image. Inspired by the fact that in many applications, such as parameter selection, a series of distorted images are available, the authors propose a novel comparison-based image quality assessment (C-IQA) method. The new comparison-based framework parallels FR-IQA by requiring two input images, and resembles NR-IQA by not using the original image. As a result, the new comparison-based approach has more application scenarios than FR-IQA does, and takes greater advantage of the accessible information than the traditional single-input NR-IQA does. Further, C-IQA is compared with other state-of-the-art NR-IQA methods on two widely used IQA databases. Experimental results show that C-IQA outperforms the other NR-IQA methods for parameter selection, and the parameter trimming framework combined with C-IQA saves the computation of iterative image reconstruction up to 80%.
Convolutional Neural Networks (CNNs) do not have a predictable recognition behavior with respect to the input resolution change. This prevents the feasibility of deployment on different input image resolutions for a specific model. To achieve efficient and flexible image classification at runtime, we employ meta learners to generate convolutional weights of main networks for various input scales and maintain privatized Batch Normalization layers per scale. For improved training performance, we further utilize knowledge distillation on the fly over model predictions based on different input resolutions. The learned meta network could dynamically parameterize main networks to act on input images of arbitrary size with consistently better accuracy compared to individually trained models. Extensive experiments on the ImageNet demonstrate that our method achieves an improved accuracy-efficiency trade-off during the adaptive inference process. By switching executable input resolutions, our method could satisfy the requirement of fast adaption in different resource-constrained environments. Code and models are available at https://github.com/d-li14/SAN.
Modern industry requires modern solutions for monitoring the automatic production of goods. Smart monitoring of the functionality of the mechanical parts of technology systems or machines is mandatory for a fully automatic production process. Although Deep Learning has been advancing, allowing for real-time object detection and other tasks, little has been investigated about the effectiveness of specially designed Convolutional Neural Networks for defect detection and industrial object recognition. In the particular study, we employed six publically available industrial-related datasets containing defect materials and industrial tools or engine parts, aiming to develop a specialized model for pattern recognition. Motivated by the recent success of the Virtual Geometry Group (VGG) network, we propose a modified version of it, called Multipath VGG19, which allows for more local and global feature extraction, while the extra features are fused via concatenation. The experiments verified the effectiveness of MVGG19 over the traditional VGG19. Specifically, top classification performance was achieved in five of the six image datasets, while the average classification improvement was 6.95%.
In image set classification, a considerable progress has been made by representing original image sets on Grassmann manifolds. In order to extend the advantages of the Euclidean based dimensionality reduction methods to the Grassmann Manifold, several methods have been suggested recently which jointly perform dimensionality reduction and metric learning on Grassmann manifold to improve performance. Nevertheless, when applied to complex datasets, the learned features do not exhibit enough discriminatory power. To overcome this problem, we propose a new method named Grassmannian Discriminant Maps (GDM) for manifold dimensionality reduction problems. The core of the method is a new discriminant function for metric learning and dimensionality reduction. For comparison and better understanding, we also study a simple variations to GDM. The key difference between them is the discriminant function. We experiment on data sets corresponding to three tasks: face recognition, object categorization, and hand gesture recognition to evaluate the proposed method and its simple extensions. Compared with the state of the art, the results achieved show the effectiveness of the proposed algorithm.
We treat the problem of color enhancement as an image translation task, which we tackle using both supervised and unsupervised learning. Unlike traditional image to image generators, our translation is performed using a global parameterized color transformation instead of learning to directly map image information. In the supervised case, every training image is paired with a desired target image and a convolutional neural network (CNN) learns from the expert retouched images the parameters of the transformation. In the unpaired case, we employ two-way generative adversarial networks (GANs) to learn these parameters and apply a circularity constraint. We achieve state-of-the-art results compared to both supervised (paired data) and unsupervised (unpaired data) image enhancement methods on the MIT-Adobe FiveK benchmark. Moreover, we show the generalization capability of our method, by applying it on photos from the early 20th century and to dark video frames.
An image is worth a thousand words, conveying information that goes beyond the mere visual content therein. In this paper, we study the intent behind social media images with an aim to analyze how visual information can facilitate recognition of human intent. Towards this goal, we introduce an intent dataset, Intentonomy, comprising 14K images covering a wide range of everyday scenes. These images are manually annotated with 28 intent categories derived from a social psychology taxonomy. We then systematically study whether, and to what extent, commonly used visual information, i.e., object and context, contribute to human motive understanding. Based on our findings, we conduct further study to quantify the effect of attending to object and context classes as well as textual information in the form of hashtags when training an intent classifier. Our results quantitatively and qualitatively shed light on how visual and textual information can produce observable effects when predicting intent.
Open-world learning is a problem where an autonomous agent detects things that it does not know and learns them over time from a non-stationary and never-ending stream of data; in an open-world environment, the training data and objective criteria are never available at once. The agent should grasp new knowledge from learning without forgetting acquired prior knowledge. Researchers proposed a few open-world learning agents for image classification tasks that operate in complex scenarios. However, all prior work on open-world learning has all labeled data to learn the new classes from the stream of images. In scenarios where autonomous agents should respond in near real-time or work in areas with limited communication infrastructure, human labeling of data is not possible. Therefore, supervised open-world learning agents are not scalable solutions for such applications. Herein, we propose a new framework that enables agents to learn new classes from a stream of unlabeled data in an unsupervised manner. Also, we study the robustness and learning speed of such agents with supervised and unsupervised feature representation. We also introduce a new metric for open-world learning without labels. We anticipate our theories and method to be a starting point for developing autonomous true open-world never-ending learning agents.
Adversarial examples have shown that albeit highly accurate, models learned by machines, differently from humans,have many weaknesses. However, humans' perception is also fundamentally different from machines, because we do not see the signals which arrive at the retina but a rather complex recreation of them. In this paper, we explore how machines could recreate the input as well as investigate the benefits of such an augmented perception. In this regard, we propose Perceptual Deep Neural Networks ($\varphi$DNN) which also recreate their own input before further processing. The concept is formalized mathematically and two variations of it are developed (one based on inpainting the whole image and the other based on a noisy resized super resolution recreation). Experiments reveal that $\varphi$DNNs can reduce attacks' accuracy substantially, surpassing state-of-the-art defenses in 87% of the tests for adversarial training variations and 100% of the tests when only comparing with other pre-processing type of defenses. Moreover, the recreation process intentionally corrupts the input image. Interestingly, we show by ablation tests that corrupting the input is, although counter-intuitive,beneficial. This suggests that the blind-spot in vertebrates might also be, analogously, the precursor of visual robustness. Thus, $\varphi$DNNs reveal that input recreation has strong benefits for artificial neural networks similar to biological ones, shedding light into the importance of the blind-spot and starting an area of perception models for robust recognition in artificial intelligence.
Hyperspectral image (HSI) classification is widely used for the analysis of remotely sensed images. Hyperspectral imagery includes varying bands of images. Convolutional Neural Network (CNN) is one of the most frequently used deep learning based methods for visual data processing. The use of CNN for HSI classification is also visible in recent works. These approaches are mostly based on 2D CNN. Whereas, the HSI classification performance is highly dependent on both spatial and spectral information. Very few methods have utilized the 3D CNN because of increased computational complexity. This letter proposes a Hybrid Spectral Convolutional Neural Network (HybridSN) for HSI classification. Basically, the HybridSN is a spectral-spatial 3D-CNN followed by spatial 2D-CNN. The 3D-CNN facilitates the joint spatial-spectral feature representation from a stack of spectral bands. The 2D-CNN on top of the 3D-CNN further learns more abstract level spatial representation. Moreover, the use of hybrid CNNs reduces the complexity of the model compared to 3D-CNN alone. To test the performance of this hybrid approach, very rigorous HSI classification experiments are performed over Indian Pines, Pavia University and Salinas Scene remote sensing datasets. The results are compared with the state-of-the-art hand-crafted as well as end-to-end deep learning based methods. A very satisfactory performance is obtained using the proposed HybridSN for HSI classification. The source code can be found at \url{https://github.com/gokriznastic/HybridSN}.
This paper strives to find the sentence best describing the content of an image or video. Different from existing works, which rely on a joint subspace for image / video to sentence matching, we propose to do so in a visual space only. We contribute Word2VisualVec, a deep neural network architecture that learns to predict a deep visual encoding of textual input based on sentence vectorization and a multi-layer perceptron. We thoroughly analyze its architectural design, by varying the sentence vectorization strategy, network depth and the deep feature to predict for image to sentence matching. We also generalize Word2VisualVec for matching a video to a sentence, by extending the predictive abilities to 3-D ConvNet features as well as a visual-audio representation. Experiments on four challenging image and video benchmarks detail Word2VisualVec's properties, capabilities for image and video to sentence matching, and on all datasets its state-of-the-art results.