Abstract: Given the often enormous effort required to train GANs, both computationally as well as in dataset collection, the re-use of pretrained GANs largely increases the potential impact of generative models. Therefore, we propose a novel knowledge transfer method for generative models based on mining the knowledge that is most beneficial to a specific target domain, either from a single or multiple pretrained GANs. This is done using a miner network that identifies which part of the generative distribution of each pretrained GAN outputs samples closest to the target domain. Mining effectively steers GAN sampling towards suitable regions of the latent space, which facilitates the posterior finetuning and avoids pathologies of other methods, such as mode collapse and lack of flexibility. Furthermore, to prevent overfitting on small target domains, we introduce sparse subnetwork selection, that restricts the set of trainable neurons to those that are relevant for the target dataset. We perform comprehensive experiments on several challenging datasets using various GAN architectures (BigGAN, Progressive GAN, and StyleGAN) and show that the proposed method, called MineGAN, effectively transfers knowledge to domains with few target images, outperforming existing methods. In addition, MineGAN can successfully transfer knowledge from multiple pretrained GANs. MineGAN.

Abstract: In this work, we study the suitability of high dynamic range, wide color gamut (HDR/WCG) objective quality metrics to assess the perceived deterioration of compressed images encoded using the hybrid log-gamma (HLG) method, which is the standard for HDR television. Several image quality metrics have been developed to deal specifically with HDR content, although in previous work we showed that the best results (i.e., better matches to the opinion of human expert observers) are obtained by an HDR metric that consists simply in applying a given standard dynamic range metric, called visual information fidelity (VIF), directly to HLG-encoded images. However, all these HDR metrics ignore the chroma components for their calculations, that is, they consider only the luminance channel. For this reason, in the current work, we conduct subjective evaluation experiments in a professional setting using compressed HDR/WCG images encoded with HLG and analyze the ability of the best HDR metric to detect perceivable distortions in the chroma components, as well as the suitability of popular color metrics (including ΔITPR , which supports parameters for HLG) to correlate with the opinion scores. Our first contribution is to show that there is a need to consider the chroma components in HDR metrics, as there are color distortions that subjects perceive but that the best HDR metric fails to detect. Our second contribution is the surprising result that VIF, which utilizes only the luminance channel, correlates much better with the subjective evaluation scores than the metrics investigated that do consider the color components.

Abstract: Deep video action recognition models have been highly successful in recent years but require large quantities of manually-annotated data, which are expensive and laborious to obtain. In this work, we investigate the generation of synthetic training data for video action recognition, as synthetic data have been successfully used to supervise models for a variety of other computer vision tasks. We propose an interpretable parametric generative model of human action videos that relies on procedural generation, physics models and other components of modern game engines. With this model we generate a diverse, realistic, and physically plausible dataset of human action videos, called PHAV for “Procedural Human Action Videos”. PHAV contains a total of 39,982 videos, with more than 1000 examples for each of 35 action categories. Our video generation approach is not limited to existing motion capture sequences: 14 of these 35 categories are procedurally-defined synthetic actions. In addition, each video is represented with 6 different data modalities, including RGB, optical flow and pixel-level semantic labels. These modalities are generated almost simultaneously using the Multiple Render Targets feature of modern GPUs. In order to leverage PHAV, we introduce a deep multi-task (i.e. that considers action classes from multiple datasets) representation learning architecture that is able to simultaneously learn from synthetic and real video datasets, even when their action categories differ. Our experiments on the UCF-101 and HMDB-51 benchmarks suggest that combining our large set of synthetic videos with small real-world datasets can boost recognition performance. Our approach also significantly outperforms video representations produced by fine-tuning state-of-the-art unsupervised generative models of videos.

Abstract: Image retrieval can be dealt by combining standard descriptors, such as those of MPEG-7, which are defined independently for each visual cue (e.g. SCD or CLD for Color, HTD for texture or EHD for edges).
A common problem is to combine similarities coming from descriptors representing different concepts in different spaces. In this paper we propose a color texture description that bypasses this problem from its inherent definition. It is based on a low dimensional space with 6 perceptual axes. Texture is described in a 3D space derived from a direct implementation of the original Julesz’s Texton theory and color is described in a 3D perceptual space. This early fusion through the blob concept in these two bounded spaces avoids the problem and allows us to derive a sparse color-texture descriptor that achieves similar performance compared to MPEG-7 in image retrieval. Moreover, our descriptor presents comprehensive qualities since it can also be applied either in segmentation or browsing: (a) a dense image representation is defined from the descriptor showing a reasonable performance in locating texture patterns included in complex images; and (b) a vocabulary of basic terms is derived to build an intermediate level descriptor in natural language improving browsing by bridging semantic gap

Abstract: Bag-of-words based image representation is a successful approach for object recognition. Generally, the subsequent stages of the process: feature detection,feature description, vocabulary construction and image representation are performed independent of the intentioned object classes to be detected. In such a framework, it was found that the combination of different image cues, such as shape and color, often obtains below expected results. This paper presents a novel method for recognizing object categories when using ultiple cues by separately processing the shape and color cues and combining them by modulating the shape features by category specific color attention. Color is used to compute bottom up and top-down attention maps. Subsequently, these color attention maps are used to modulate the weights of the shape features. In regions with higher attention shape features are given more weight than in regions with low attention. We compare our approach with existing methods that combine color and shape cues on five data sets containing varied importance of both cues, namely, Soccer (color predominance), Flower (color and hape parity), PASCAL VOC 2007 and 2009 (shape predominance) and Caltech-101 (color co-interference). The experiments clearly demonstrate that in all five data sets our proposed framework significantly outperforms existing methods for combining color and shape information.

Abstract: This paper investigates the role of saliency to improve the classification accuracy of a Convolutional Neural Network (CNN) for the case when scarce training data is available. Our approach consists in adding a saliency branch to an existing CNN architecture which is used to modulate the standard bottom-up visual features from the original image input, acting as an attentional mechanism that guides the feature extraction process. The main aim of the proposed approach is to enable the effective training of a fine-grained recognition model with limited training samples and to improve the performance on the task, thereby alleviating the need to annotate a large dataset. The vast majority of saliency methods are evaluated on their ability to generate saliency maps, and not on their functionality in a complete vision pipeline. Our proposed pipeline allows to evaluate saliency methods for the high-level task of object recognition. We perform extensive experiments on various fine-grained datasets (Flowers, Birds, Cars, and Dogs) under different conditions and show that saliency can considerably improve the network’s performance, especially for the case of scarce training data. Furthermore, our experiments show that saliency methods that obtain improved saliency maps (as measured by traditional saliency benchmarks) also translate to saliency methods that yield improved performance gains when applied in an object recognition pipeline.

Abstract: The gamut mapping algorithm is one of the most promising methods to achieve computational color constancy. However, so far, gamut mapping algorithms are restricted to the use of pixel values to estimate the illuminant. Therefore, in this paper, gamut mapping is extended to incorporate the statistical nature of images. It is analytically shown that the proposed gamut mapping framework is able to include any linear filter output. The main focus is on the local n-jet describing the derivative structure of an image. It is shown that derivatives have the advantage over pixel values to be invariant to disturbing effects (i.e. deviations of the diagonal model) such as saturated colors and diffuse light. Further, as the n-jet based gamut mapping has the ability to use more information than pixel values alone, the combination of these algorithms are more stable than the regular gamut mapping algorithm. Different methods of combining are proposed. Based on theoretical and experimental results conducted on large scale data sets of hyperspectral, laboratory and realworld scenes, it can be derived that (1) in case of deviations of the diagonal model, the derivative-based approach outperforms the pixel-based gamut mapping, (2) state-of-the-art algorithms are outperformed by the n-jet based gamut mapping, (3) the combination of the different n-jet based gamut