Abstract:Creating a trailer requires carefully picking out and piecing together brief enticing moments out of a longer video, making it a challenging and time-consuming task. This requires selecting moments based on both visual and dialogue information. We introduce a multi-modal method for predicting the trailerness to assist editors in selecting trailer-worthy moments from long-form videos. We present results on a newly introduced soap opera dataset, demonstrating that predicting trailerness is a challenging task that benefits from multi-modal information. Code is available at https://github.com/carlobretti/cliffhanger
Abstract:This paper explores the capacity of computer vision models to discern temporal information in visual content, focusing specifically on historical photographs. We investigate the dating of images using OpenCLIP, an open-source implementation of CLIP, a multi-modal language and vision model. Our experiment consists of three steps: zero-shot classification, fine-tuning, and analysis of visual content. We use the \textit{De Boer Scene Detection} dataset, containing 39,866 gray-scale historical press photographs from 1950 to 1999. The results show that zero-shot classification is relatively ineffective for image dating, with a bias towards predicting dates in the past. Fine-tuning OpenCLIP with a logistic classifier improves performance and eliminates the bias. Additionally, our analysis reveals that images featuring buses, cars, cats, dogs, and people are more accurately dated, suggesting the presence of temporal markers. The study highlights the potential of machine learning models like OpenCLIP in dating images and emphasizes the importance of fine-tuning for accurate temporal analysis. Future research should explore the application of these findings to color photographs and diverse datasets.
Abstract:Dataset summarisation is a fruitful approach to dataset inspection. However, when applied to a single dataset the discovery of visual concepts is restricted to those most prominent. We argue that a comparative approach can expand upon this paradigm to enable richer forms of dataset inspection that go beyond the most prominent concepts. To enable dataset comparison we present a module that learns concept-level prototypes across datasets. We leverage self-supervised learning to discover these prototypes without supervision, and we demonstrate the benefits of our approach in two case-studies. Our findings show that dataset comparison extends dataset inspection and we hope to encourage more works in this direction. Code and usage instructions available at https://github.com/Nanne/ProtoSim
Abstract:We propose Hierarchical ProtoPNet: an interpretable network that explains its reasoning process by considering the hierarchical relationship between classes. Different from previous methods that explain their reasoning process by dissecting the input image and finding the prototypical parts responsible for the classification, we propose to explain the reasoning process for video action classification by dissecting the input video frames on multiple levels of the class hierarchy. The explanations leverage the hierarchy to deal with uncertainty, akin to human reasoning: When we observe water and human activity, but no definitive action it can be recognized as the water sports parent class. Only after observing a person swimming can we definitively refine it to the swimming action. Experiments on ActivityNet and UCF-101 show performance improvements while providing multi-level explanations.
Abstract:There is a bidirectional relationship between culture and AI; AI models are increasingly used to analyse culture, thereby shaping our understanding of culture. On the other hand, the models are trained on collections of cultural artifacts thereby implicitly, and not always correctly, encoding expressions of culture. This creates a tension that both limits the use of AI for analysing culture and leads to problems in AI with respect to cultural complex issues such as bias. One approach to overcome this tension is to more extensively take into account the intricacies and complexities of culture. We structure our discussion using four concepts that guide humanistic inquiry into culture: subjectivity, scalability, contextuality, and temporality. We focus on these concepts because they have not yet been sufficiently represented in AI research. We believe that possible implementations of these aspects into AI research leads to AI that better captures the complexities of culture. In what follows, we briefly describe these four concepts and their absence in AI research. For each concept, we define possible research challenges.
Abstract:For image segmentation, the current standard is to perform pixel-level optimization and inference in Euclidean output embedding spaces through linear hyperplanes. In this work, we show that hyperbolic manifolds provide a valuable alternative for image segmentation and propose a tractable formulation of hierarchical pixel-level classification in hyperbolic space. Hyperbolic Image Segmentation opens up new possibilities and practical benefits for segmentation, such as uncertainty estimation and boundary information for free, zero-label generalization, and increased performance in low-dimensional output embeddings.
Abstract:E-commerce provides rich multimodal data that is barely leveraged in practice. One aspect of this data is a category tree that is being used in search and recommendation. However, in practice, during a user's session there is often a mismatch between a textual and a visual representation of a given category. Motivated by the problem, we introduce the task of category-to-image retrieval in e-commerce and propose a model for the task, CLIP-ITA. The model leverages information from multiple modalities (textual, visual, and attribute modality) to create product representations. We explore how adding information from multiple modalities (textual, visual, and attribute modality) impacts the model's performance. In particular, we observe that CLIP-ITA significantly outperforms a comparable model that leverages only the visual modality and a comparable model that leverages the visual and attribute modality.
Abstract:Visual Place Recognition (VPR) is generally concerned with localizing outdoor images. However, localizing indoor scenes that contain part of an outdoor scene can be of large value for a wide range of applications. In this paper, we introduce Inside Out Visual Place Recognition (IOVPR), a task aiming to localize images based on outdoor scenes visible through windows. For this task we present the new large-scale dataset Amsterdam-XXXL, with images taken in Amsterdam, that consists of 6.4 million panoramic street-view images and 1000 user-generated indoor queries. Additionally, we introduce a new training protocol Inside Out Data Augmentation to adapt Visual Place Recognition methods for localizing indoor images, demonstrating the potential of Inside Out Visual Place Recognition. We empirically show the benefits of our proposed data augmentation scheme on a smaller scale, whilst demonstrating the difficulty of this large-scale dataset for existing methods. With this new task we aim to encourage development of methods for IOVPR. The dataset and code are available for research purposes at https://github.com/saibr/IOVPR
Abstract:The Synesthetic Variational Autoencoder (SynVAE) introduced in this research is able to learn a consistent mapping between visual and auditive sensory modalities in the absence of paired datasets. A quantitative evaluation on MNIST as well as the Behance Artistic Media dataset (BAM) shows that SynVAE is capable of retaining sufficient information content during the translation while maintaining cross-modal latent space consistency. In a qualitative evaluation trial, human evaluators were furthermore able to match musical samples with the images which generated them with accuracies of up to 73%.
Abstract:Adversarial training has been recently employed for realizing structured semantic segmentation, in which the aim is to preserve higher-level scene structural consistencies in dense predictions. However, as we show, value-based discrimination between the predictions from the segmentation network and ground-truth annotations can hinder the training process from learning to improve structural qualities as well as disabling the network from properly expressing uncertainties. In this paper, we rethink adversarial training for semantic segmentation and propose to formulate the fake/real discrimination framework with a correct/incorrect training objective. More specifically, we replace the discriminator with a "gambler" network that learns to spot and distribute its budget in areas where the predictions are clearly wrong, while the segmenter network tries to leave no clear clues for the gambler where to bet. Empirical evaluation on two road-scene semantic segmentation tasks shows that not only does the proposed method re-enable expressing uncertainties, it also improves pixel-wise and structure-based metrics.