We propose a novel strategy, ES3, for self-supervised learning of robust audio-visual speech representations from unlabeled talking face videos. While many recent approaches for this task primarily rely on guiding the learning process using the audio modality alone to capture information shared between audio and video, we reframe the problem as the acquisition of shared, unique (modality-specific) and synergistic speech information to address the inherent asymmetry between the modalities. Based on this formulation, we propose a novel “evolving” strategy that progressively builds joint audio-visual speech representations that are strong for both uni-modal (audio & visual) and bi-modal (audio-visual) speech. First, we leverage the more easily learnable audio modality to initialize audio and visual representations by capturing audio-unique and shared speech information. Next, we incorporate video-unique speech information and bootstrap the audio-visual repre-sentations on top of the previously acquired shared knowledge. Finally, we maximize the total audio-visual speech information, including synergistic information to obtain robust and comprehensive representations. We implement ES3as a simple Siamese framework and experiments on both English benchmarks and a newly contributed large-scale Mandarin dataset show its effectiveness. In particular, on LRS2-BBC, our smallest model is on par with SoTA models with only 1/2 parameters and 1/8 unlabeled data (223h).

We propose a new efficient framework, the Unified Context Network (UniCon), for robust active speaker detection (ASD). Traditional methods for ASD usually operate on each candidate's pre-cropped face track separately and do not sufficiently consider the relationships among the candidates. This potentially limits performance, especially in challenging scenarios with low-resolution faces, multiple candidates, etc. Our solution is a novel, unified framework that focuses on jointly modeling multiple types of contextual information: spatial context to indicate the position and scale of each candidate's face, relational context to capture the visual relationships among the candidates and contrast audio-visual affinities with each other, and temporal context to aggregate long-term information and smooth out local uncertainties. Based on such information, our model optimizes all candidates in a unified process for robust and reliable ASD. A thorough ablation study is performed on several challenging ASD benchmarks under different settings. In particular, our method outperforms the state-of-the-art by a large margin of about 15% mean Average Precision (mAP) absolute on two challenging subsets: one with three candidate speakers, and the other with faces smaller than 64 pixels. Together, our UniCon achieves 92.0% mAP on the AVA-ActiveSpeaker validation set, surpassing 90% for the first time on this challenging dataset at the time of submission. Project website: https://unicon-asd.github.io/.

Lip reading is the task of recognizing speech content by analyzing movements in the lip region when people are speaking. Based on the continuity in adjacent frames in the speaking process, and the consistency in motion patterns among different people when they pronounce the same phoneme, we model lip movements as a sequence of apparent deformations in the lip region during the speaking process. Specifically, we introduce a Deformation Flow Network (DFN) to learn the deformation flow between adjacent frames, which directly captures the motion information within the lip region. The learned deformation flow is then combined with the original grayscale frames with a two-stream network to perform lip reading. To make the two streams learn from each other in the learning process, we introduce a bidirectional knowledge distillation loss to train the two branches jointly. Owing to the complementary cues provided by different branches, the two-stream network shows substantial improvement over using either single branch. A thorough experimental evaluation on two large-scale lip reading benchmarks is presented with detailed analysis. The results accord with our motivation, and show that our method achieves state-of-the-art or comparable performance on these two challenging datasets.

In this paper, we propose a multi-modal multi-feature (M$^{3}F$) approach for in-the-wild valence-arousal estimation. In the proposed M$^{3}F$ framework, we fuse both visual features from videos and acoustic features from the audio tracks to estimate the valence and arousal. We follow a CNN-RNN paradigm, where the spatio-temporal visual features are extracted with a 3D convolutional network and/or a pretrained 2D convolutional network, and a bidirectional recurrent neural network. We evaluated the M$^{3}F$ framework on the validation set provided by the Affective Behavior Analysis in-the-wild (ABAW) Challenge, held in conjunction with the IEEE International Conference on Automatic Face and Gesture Recognition (FG) 2020, and it significantly outperforms the baseline method.

Recent advances in deep learning have heightened interest among researchers in the field of visual speech recognition (VSR). Currently, most existing methods equate VSR with automatic lip reading, which attempts to recognise speech by analysing lip motion. However, human experience and psychological studies suggest that we do not always fix our gaze at each other's lips during a face-to-face conversation, but rather scan the whole face repetitively. This inspires us to revisit a fundamental yet somehow overlooked problem: can VSR models benefit from reading extraoral facial regions, i.e. beyond the lips? In this paper, we perform a comprehensive study on the evaluation of the effects of different facial regions with state-of-the-art VSR models, including the mouth, the whole face, the upper face, and even the cheeks. Experiments are conducted on both word-level and sentence-level benchmarks with different characteristics. We find that despite the complex variations of the data, incorporating information from extraoral facial regions, even the upper face, consistently benefits VSR performance. Furthermore, we introduce a simple yet effective method based on Cutout to learn more discriminative features for face-based VSR, hoping to maximise the utility of information encoded in different facial regions. Our experiments show obvious improvements over existing state-of-the-art methods that use only the lip region as inputs, a result we believe would probably provide the VSR community with some new and exciting insights.

Large-scale datasets have successively proven their fundamental importance in several research fields, especially for early progress in some emerging topics. In this paper, we focus on the problem of visual speech recognition, also known as lip-reading, which has received increasing interest in recent years. We present a naturally-distributed large-scale benchmark for lip-reading in the wild, named LRW-1000, which contains 1,000 classes with 718,018 samples from more than 2,000 individual speakers. Each class corresponds to the syllables of a Mandarin word composed of one or several Chinese characters. To the best of our knowledge, it is currently the largest word-level lipreading dataset and also the only public large-scale Mandarin lip-reading dataset. This dataset aims at covering a "natural" variability over different speech modes and imaging conditions to incorporate challenges encountered in practical applications. It has shown a large variation in this benchmark in several aspects, including the number of samples in each class, video resolution, lighting conditions, and speakers' attributes such as pose, age, gender, and make-up. Besides providing a detailed description of the dataset and its collection pipeline, we evaluate several typical popular lip-reading methods and perform a thorough analysis of the results from several aspects. The results demonstrate the consistency and challenges of our dataset, which may open up some new promising directions for future work.