Search

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 FROM AUTHOR]

Computer vision has established a foothold in the online fashion retail industry. Main product detection is a crucial step of vision-based fashion product feed parsing pipelines, focused on identifying the bounding boxes that contain the product being sold in the gallery of images of the product page. The current state-of-the-art approach does not leverage the relations between regions in the image, and treats images of the same product independently, therefore not fully exploiting visual and product contextual information. In this paper, we propose a model that incorporates Graph Convolutional Networks (GCN) that jointly represent all detected bounding boxes in the gallery as nodes. We show that the proposed method is better than the state-of-the-art, especially, when we consider the scenario where title-input is missing at inference time and for cross-dataset evaluation, our method outperforms previous approaches by a large margin. [ABSTRACT FROM AUTHOR]

The particular properties of argumental compounds in Italian pose interesting theoretical challenges, and investigations of possible syntactic operations within this type of complex words have resulted in conflicting conclusions. Regarding compound-internal anaphora, some researchers exclude the possibility that pronouns can refer to the non-head, while others do not. However, these findings have been based on researchers' intuitions and on occurrences in language corpora, and while intuitions have been shown to give contrasting results, the absence of a grammatical structure in a corpus should not be taken as evidence that the structure is not possible. The present study aims to experimentally determine the possibility of compound-internal pronominal reference based on structural properties of compounds and referential expressions. Judgements were obtained from 140 Italian native speakers who rated the acceptability of sentences containing a pronoun (null or overt) referring to the argument element of an argumental compound. The results indicate that compound-internal anaphoric reference is acceptable in the case of left-headed compounds and, to a somewhat lesser extent, of verb-noun compounds. The argument element of right-headed compounds, however, does not appear to be available to anaphoric reference. Referential expressions also play a role in the degree of acceptability, with left-headed compounds allowing null form anaphora to a greater extent. These results provide new evidence on compound-internal pronominal reference and give important insights into the processing of argumental compounds. [ABSTRACT FROM AUTHOR]

Every day we make choices, but our degree of investment in them differs, both in terms of pre-verbal experience and verbal justification. In an earlier experimental study, participants were asked to pick the more attractive one among two human faces, and among two abstract figures, and later to provide verbal motivations for these choices. They did not know that in some of the cases their choices were manipulated (i.e., they were asked to motivate the item they had not chosen). Against claims about our unreliability as conscious agents (Nisbett et al. in Psychol Rev 84:231–259, 2013; Johansson et al. in Science 310:116–119, 2005), the study found that in about half the cases the manipulations were detected. In the present study, we investigated whether varying degrees of choice investment could be an explanatory factor for such findings. We analysed the verbal justifications of the participants along a set of semantic categories, based on theoretical ideas from phenomenology and cognitive linguistics, and formulated a matrix of eleven markers of choice investment. We predicted a greater degree of investment when motivating (a) choices of faces than figures, (b) manipulated than actual choices, and (c) detected than non-detected manipulations. These predictions were confirmed, but with various strength. This allows us to argue for both consilience and differences between pre-verbal choice investment and the corresponding verbal motivations of the choices made, and thus for (degrees of) conscious awareness of choice making. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of inferring unseen cross-modal image-to-image translations between multiple modalities. We assume that only some of the pairwise translations have been seen (i.e. trained) and infer the remaining unseen translations (where training pairs are not available). We propose mix and match networks, an approach where multiple encoders and decoders are aligned in such a way that the desired translation can be obtained by simply cascading the source encoder and the target decoder, even when they have not interacted during the training stage (i.e. unseen). The main challenge lies in the alignment of the latent representations at the bottlenecks of encoder–decoder pairs. We propose an architecture with several tools to encourage alignment, including autoencoders and robust side information and latent consistency losses. We show the benefits of our approach in terms of effectiveness and scalability compared with other pairwise image-to-image translation approaches. We also propose zero-pair cross-modal image translation, a challenging setting where the objective is inferring semantic segmentation from depth (and vice-versa) without explicit segmentation-depth pairs, and only from two (disjoint) segmentation-RGB and depth-RGB training sets. We observe that a certain part of the shared information between unseen modalities might not be reachable, so we further propose a variant that leverages pseudo-pairs which allows us to exploit this shared information between the unseen modalities. [ABSTRACT FROM AUTHOR]

Salient object segmentation in videos is generally broken up in a video segmentation part and a saliency assignment part. Recently, object proposals, which are used to segment the image, have had significant impact on many computer vision applications, including image segmentation, object detection, and recently saliency detection in still images. However, their usage has not yet been evaluated for salient object segmentation in videos. Therefore, in this paper, we investigate the application of object proposals to salient object segmentation in videos. In addition, we propose a new motion feature derived from the optical flow structure tensor for video saliency detection. Experiments on two standard benchmark datasets for video saliency show that the proposed motion feature improves saliency estimation results, and that object proposals are an efficient method for salient object segmentation. Results on the challenging SegTrack v2 and Fukuchi benchmark data sets show that we significantly outperform the state-of-the-art. [ABSTRACT FROM AUTHOR]

In emergency situations, actions that save lives and limit the impact of hazards are crucial. In order to act, situational awareness is needed to decide what to do. Geolocalized photos and video of the situations as they evolve can be crucial in better understanding them and making decisions faster. Cameras are almost everywhere these days, either in terms of smartphones, installed CCTV cameras, UAVs or others. However, this poses challenges in big data and information overflow. Moreover, most of the time there are no disasters at any given location, so humans aiming to detect sudden situations may not be as alert as needed at any point in time. Consequently, computer vision tools can be an excellent decision support. The number of emergencies where computer vision tools has been considered or used is very wide, and there is a great overlap across related emergency research. Researchers tend to focus on state-of-the-art systems that cover the same emergency as they are studying, obviating important research in other fields. In order to unveil this overlap, the survey is divided along four main axes: the types of emergencies that have been studied in computer vision, the objective that the algorithms can address, the type of hardware needed and the algorithms used. Therefore, this review provides a broad overview of the progress of computer vision covering all sorts of emergencies. [ABSTRACT FROM AUTHOR]

Color description is one of the fundamental problems of image understanding. One of the popular ways to represent colors is by means of color names. Most existing work on color names focuses on only the eleven basic color terms of the English language. This could be limiting the discriminative power of these representations, and representations based on more color names are expected to perform better. However, there exists no clear strategy to choose additional color names. We collect a dataset of 28 additional color names. To ensure that the resulting color representation has high discriminative power we propose a method to order the additional color names according to their complementary nature with the basic color names. This allows us to compute color name representations with high discriminative power of arbitrary length. In the experiments we show that these new color name descriptors outperform the existing color name descriptor on the task of visual tracking, person re-identification and image classification. [ABSTRACT FROM AUTHOR]

Most approaches to human attribute and action recognition in still images are based on image representation in which multi-scale local features are pooled across scale into a single, scale-invariant encoding. Both in bag-of-words and the recently popular representations based on convolutional neural networks, local features are computed at multiple scales. However, these multi-scale convolutional features are pooled into a single scale-invariant representation. We argue that entirely scale-invariant image representations are sub-optimal and investigate approaches to scale coding within a bag of deep features framework. Our approach encodes multi-scale information explicitly during the image encoding stage. We propose two strategies to encode multi-scale information explicitly in the final image representation. We validate our two scale coding techniques on five datasets: Willow, PASCAL VOC 2010, PASCAL VOC 2012, Stanford-40 and Human Attributes (HAT-27). On all datasets, the proposed scale coding approaches outperform both the scale-invariant method and the standard deep features of the same network. Further, combining our scale coding approaches with standard deep features leads to consistent improvement over the state of the art. [ABSTRACT FROM AUTHOR]