Abstract

Learning an efficient label embedding framework for word images enables effective word spotting of handwritten documents. In this work, we propose different schemes of label embedding for word images using deep neural architectures and their representations. We refer to our first scheme as the two-stage label embedding technique which projects both word images and their corresponding textual strings into a common subspace. We further introduce an end-to-end label embedding scheme using deep neural architecture which simplifies the embedding process and reports state-of-the-art performance for the task of word spotting and recognition. We also validate the role of synthetic data as a complementary modality to further enhance the embedding process. On the challenging IAM handwritten dataset, we report an mAP of 0.9753 for query-by-string-based word spotting, while under lexicon-based word recognition, our proposed method reports 1.67 and 3.62 character and word error rates, respectively. We also present the detailed ablation study on various variants of our end-to-end embedding architecture and perform analysis under varying embedding sizes. We further validate the embedding scheme on degraded printed document datasets from both Latin and Indic scripts.

Abstract

Talking-face video generation works have achieved state-of-the-art results in synthesizing videos with lip synchronization. However, most of the previous works deal with low-resolution talking-face videos (up to 256×256 pixels), thus, generating extremely high-resolution videos still remains a challenge. We take a giant leap in this work and propose a novel method to synthesize talking-face videos at resolutions as high as 4K! Our task presents several key challenges: (i) Scaling the existing methods to such high resolutions is resource-constrained, both in terms of compute and the availability of very high-resolution datasets, (ii) The synthesized videos need to be spatially and temporally coherent. The sheer number of pixels that the model needs to generate while maintaining the temporal consistency at the video level makes this task non-trivial and has never been attempted before in literature. To address these issues, we propose to train the lip-sync generator in a compact Vector Quantized (VQ) space for the first time. Our

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

Over the last decade, online lecture videos have become increasingly popular and have experienced a meteoric rise during the pandemic. However, video-language research has primarily focused on instructional videos or movies, and tools to help students navigate the growing online lectures are lacking. Our first contribution is to facilitate research in the educational domain by introducing AVLectures, a large-scale dataset consisting of 86 courses with over 2,350 lectures covering various STEM subjects. Each course contains video lectures, transcripts, OCR outputs for lecture frames, and optionally lecture notes, slides, assignments, and related educational content that can inspire a variety of tasks. Our second contribution is introducing video lecture segmentation that splits lectures into bitesized topics. Lecture clip representations leverage visual, textual, and OCR cues and are trained on a pretext selfsupervised task of matching the narration with the temporally aligned visual content. We formulate lecture segmentation as an unsupervised task and use these representations to generate segments using a temporally consistent 1- nearest neighbor algorithm, TW-FINCH [44]. We evaluate our method on 15 courses and compare it against various visual and textual baselines, outperforming all of them. Our comprehensive ablation studies also identify the key factors driving the success of our approach.

Abstract

Autonomous driving and assistance systems rely on annotated data from traffic and road scenarios to model and learn the various object relations in complex real-world scenarios. Preparation and training of deploy-able deep learning architectures require the models to be suited to different traffic scenarios and adapt to different situations. Currently, existing datasets, while large-scale, lack such diversities and are geographically biased towards mainly developed cities. An unstructured and complex driving layout found in several developing countries such as India poses a challenge to these models due to the sheer degree of variations in the object types, densities, and locations. To facilitate better research toward accommodating such scenarios, we build a new dataset, IDD-3D, which consists of multimodal data from multiple cameras and LiDAR sensors with 12k annotated driving LiDAR frames across various traffic scenarios. We discuss the need for this dataset through statistical comparisons with existing datasets and highlight benchmarks on standard 3D object detection and tracking tasks in complex layouts. Code and data available 1 .

Abstract

This work proposes a novel method to generate realistic talking head videos using audio and visual streams. We animate a source image by transferring head motion from a driving video using a dense motion field generated using learnable keypoints. We improve the quality of lip sync using audio as an additional input, helping the network to attend to the mouth region. We use additional priors using face segmentation and face mesh to improve the structure of the reconstructed faces. Finally, we improve the visual quality of the generations by incorporating a carefully designed identity-aware generator module. The identityaware generator takes the source image and the warped motion features as input to generate a high-quality output with fine-grained details. Our method produces state-ofthe-art results and generalizes well to unseen faces, languages, and voices. We comprehensively evaluate our approach using multiple metrics and outperforming the current techniques both qualitative and quantitatively. Our work opens up several applications, including enabling low bandwidth video calls. We release a demo video and additional information at http://cvit.iiit.ac.in/

Abstract

Annotation is a crucial first step in the semantic segmentation of urban images that facilitates the development of autonomous navigation systems. However, annotating complex urban images is time-consuming and challenging. It requires significant human effort making it expensive and error-prone. To reduce human effort during annotation, multiple images need to be annotated in a short time-span. In this paper, we introduce ScribbleNet, an interactive image segmentation algorithm to address this issue. Our approach provides users with a pre-segmented image that iteratively improves the segmentation using scribble as an annotation input. This method is based on conditional inference and exploits the learnt correlations in a deep neural network (DNN). ScribbleNet can: (1) work with urban city scenes captured in unseen environments, (2) annotate new classes not present in the training set, and (3) correct several labels at once. We compare this method with other interactive segmentation approaches on multiple datasets such as CityScapes, BDD, Mapillary Vistas, KITTI, and IDD. ScribbleNet reduces the annotation time of an image by up to 14.7 over manual annotation and up to 5.4 over the current approaches. The algorithm is integrated into the publicly available LabelMe image annotation tool and will be released as an open-source software.

Abstract

Many people with some form of hearing loss consider lipreading as their primary mode of day-to-day communication. However, finding resources to learn or improve one’s lipreading skills can be challenging. This is further exacerbated in the COVID19 pandemic due to restrictions on direct interactions with peers and speech therapists. Today, online MOOCs platforms like Coursera and Udemy have become the most effective form of training for many types of skill development. However, online lipreading resources are scarce as creating such resources is an extensive process needing months of manual effort to record hired actors. Because of the manual pipeline, such platforms are also limited in vocabulary, supported languages, accents, *Equal contribution and speakers and have a high usage cost. In this work, we investigate the possibility of replacing real human talking videos with synthetically generated videos. Synthetic data can easily incorporate larger vocabularies, variations in accent, and even local languages and many speakers. We propose an end-to-end automated pipeline to develop such a platform using state-of-the-art talking head video generator networks, text-to-speech models, and computer vision techniques. We then perform an extensive human evaluation using carefully thought out lipreading exercises to validate the quality of our designed platform against the existing lipreading platforms. Our studies concretely point toward the potential of our approach in developing a largescale lipreading MOOC platform that can impact millions of people with hearing loss

Abstract

Doubles play an indispensable role in the movie industry. They take the place of the actors in dangerous stunt scenes or scenes where the same actor plays multiple characters. The double’s face is later replaced with the actor’s face and expressions manually using expensive CGI technology, costing millions of dollars and taking months to complete. An automated, inexpensive, and fast way can be to use face-swapping techniques that aim to swap an identity from a source face video (or an image) to a target face video. However, such methods cannot preserve the source expressions of the actor important for the scene’s context. *Equal contribution To tackle this challenge, we introduce video-to-video (V2V) face-swapping, a novel task of face-swapping that can preserve (1) the identity and expressions of the source (actor) face video and (2) the background and pose of the target (double) video. We propose FaceOff, a V2V face-swapping system that operates by learning a robust blending operation to merge two face videos following the constraints above. It reduces the videos to a quantized latent space and then blends them in the reduced space. FaceOff is trained in a self-supervised manner and robustly tackles the nontrivial challenges of V2V face-swapping. As shown in the experimental section, FaceOff significantly outperforms alternate approaches qualitatively and quantitatively.

Abstract

The popularity of egocentric cameras and their always-on nature has lead to the abundance of day long first-person videos. The highly redundant nature of these videos and extreme camera-shakes make them difficult to watch from beginning to end. These videos require efficient summarization tools for consumption. However, traditional summarization techniques developed for static surveillance videos or highly curated sports videos and movies are either not suitable or simply do not scale for such hours long videos in the wild. On the other hand, specialized summarization techniques developed for egocentric videos limit their focus to important objects and people. This paper presents a novel unsupervised reinforcement learning framework to summarize egocentric videos both in terms of length and the content. The proposed framework facilitates incorporating various prior preferences such as faces, places, or …