Abstract: Observers are faster to detect a target among a set of distracters if the targets and distracters come from different color categories. This cross-boundary advantage seems to be limited to the right visual field, which is consistent with the dominance of the left hemisphere for language processing [Gilbert et al., Proc. Natl. Acad. Sci. USA 103, 489 (2006)]. Here we study whether a similar visual field advantage is found in the color identification task in speakers of Mandarin, a language that uses a logographic system. Forty late Mandarin-English bilinguals performed a blue-green color categorization task, in a blocked design, in their first language (L1: Mandarin) or second language (L2: English). Eleven color singletons ranging from blue to green were presented for 160 ms, randomly in the left visual field (LVF) or right visual field (RVF). Color boundary and reaction times (RTs) at the color boundary were estimated in L1 and L2, for both visual fields. We found that the color boundary did not differ between the languages; RTs at the color boundary, however, were on average more than 100 ms shorter in the English compared to the Mandarin sessions, but only when the stimuli were presented in the RVF. The finding may be explained by the script nature of the two languages: Mandarin logographic characters are analyzed visuospatially in the right hemisphere, which conceivably facilitates identification of color presented to the LVF.

Abstract: Multimodal learning from document data has achieved great success lately as it allows to pre-train semantically meaningful features as a prior into a learnable downstream approach. In this paper, we approach the document classification problem by learning cross-modal representations through language and vision cues, considering intra- and inter-modality relationships. Instead of merging features from different modalities into a common representation space, the proposed method exploits high-level interactions and learns relevant semantic information from effective attention flows within and across modalities. The proposed learning objective is devised between intra- and inter-modality alignment tasks, where the similarity distribution per task is computed by contracting positive sample pairs while simultaneously contrasting negative ones in the common feature representation space}. Extensive experiments on public document classification datasets demonstrate the effectiveness and the generalization capacity of our model on both low-scale and large-scale datasets.

Abstract: Video Question Answering methods focus on commonsense reasoning and visual cognition of objects or persons and their interactions over time. Current VideoQA approaches ignore the textual information present in the video. Instead, we argue that textual information is complementary to the action and provides essential contextualisation cues to the reasoning process. To this end, we propose a novel VideoQA task that requires reading and understanding the text in the video. To explore this direction, we focus on news videos and require QA systems to comprehend and answer questions about the topics presented by combining visual and textual cues in the video. We introduce the ``NewsVideoQA'' dataset that comprises more than 8,600 QA pairs on 3,000+ news videos obtained from diverse news channels from around the world. We demonstrate the limitations of current Scene Text VQA and VideoQA methods and propose ways to incorporate scene text information into VideoQA methods.

Abstract: This thesis presents several contributions to the literature of sketch based image retrieval (SBIR). In SBIR the first challenge we face is how to map two different domains to common space for effective retrieval of images, while tackling the different levels of abstraction people use to express their notion of objects around while sketching. To this extent we first propose a cross-modal learning framework that maps both sketches and text into a joint embedding space invariant to depictive style, while preserving semantics. Then we have also investigated different query types possible to encompass people's dilema in sketching certain world objects. For this we propose an approach for multi-modal image retrieval in multi-labelled images. A multi-modal deep network architecture is formulated to jointly model sketches and text as input query modalities into a common embedding space, which is then further aligned with the image feature space. This permits encoding the object-based features and its alignment with the query irrespective of the availability of the co-occurrence of different objects in the training set.
Finally, we explore the problem of zero-shot sketch-based image retrieval (ZS-SBIR), where human sketches are used as queries to conduct retrieval of photos from unseen categories. We importantly advance prior arts by proposing a novel ZS-SBIR scenario that represents a firm step forward in its practical application. The new setting uniquely recognises two important yet often neglected challenges of practical ZS-SBIR, (i) the large domain gap between amateur sketch and photo, and (ii) the necessity for moving towards large-scale retrieval. We first contribute to the community a novel ZS-SBIR dataset, QuickDraw-Extended. We also in this dissertation pave the path to the future direction of research in this domain.

Abstract: Word spotting is an important recognition task in large-scale retrieval of document collections. In most of the cases, methods are developed and evaluated assuming perfect word segmentation. In this paper, we propose an experimental framework to quantify the goodness that word segmentation has on the performance achieved by word spotting methods in identical unbiased conditions. The framework consists of generating systematic distortions on segmentation and retrieving the original queries from the distorted dataset. We have tested our framework on several established and state-of-the-art methods using George Washington and Barcelona Marriage Datasets. The experiments done allow for an estimate of the end-to-end performance of word spotting methods.

Abstract: Offline signature verification is one of the most challenging tasks in biometrics and document forensics. Unlike other verification problems, it needs to model minute but critical details between genuine and forged signatures, because a skilled falsification might often resembles the real signature with small deformation. This verification task is even harder in writer independent scenarios which is undeniably fiscal for realistic cases. In this paper, we model an offline writer independent signature verification task with a convolutional Siamese network. Siamese networks are twin networks with shared weights, which can be trained to learn a feature space where similar observations are placed in proximity. This is achieved by exposing the network to a pair of similar and dissimilar observations and minimizing the Euclidean distance between similar pairs while simultaneously maximizing it between dissimilar pairs. Experiments conducted on cross-domain datasets emphasize the capability of our network to model forgery in different languages (scripts) and handwriting styles. Moreover, our designed Siamese network, named SigNet, exceeds the state-of-the-art results on most of the benchmark signature datasets, which paves the way for further research in this direction.

Abstract: In this work we introduce a cross modal image retrieval system that allows both text and sketch as input modalities for the query. A cross-modal deep network architecture is formulated to jointly model the sketch and text input modalities as well as the the image output modality, learning a common embedding between text and images and between sketches and images. In addition, an attention model is used to selectively focus the attention on the different objects of the image, allowing for retrieval with multiple objects in the query. Experiments show that the proposed method performs the best in both single and multiple object image retrieval in standard datasets.