With the prevalence of stream media platforms serving music search and recommendation, interpreting music by understanding audio and lyrics interactively has become an important and challenging task. However, many previous works focus on refining individual components of encoder-decoder architecture mapping music to caption tokens, ignoring the potential usage of audio and lyrics correspondence. In this paper, we propose to explicitly learn the multi-modal alignment with retrieval augmentation by contrastive learning. By learning audio-lyrics correspondence, the model is guided to learn better cross-modal attention weights, thus generating high-quality caption words. We provide both theoretical and empirical results that demonstrate the advantage of the proposed method.

As one of the most important psychic stress reactions, micro-expressions (MEs), are spontaneous and transient facial expressions that can reveal the genuine emotions of human beings. Thus, recognizing MEs (MER) automatically is becoming increasingly crucial in the field of affective computing, and provides essential technical support in lie detection, psychological analysis and other areas. However, the lack of abundant ME data seriously restricts the development of cutting-edge data-driven MER models. Despite the recent efforts of several spontaneous ME datasets to alleviate this problem, it is still a tiny amount of work. To solve the problem of ME data hunger, we construct a dynamic spontaneous ME dataset with the largest current ME data scale, called DFME (Dynamic Facial Micro-expressions), which includes 7,526 well-labeled ME videos induced by 671 participants and annotated by more than 20 annotators throughout three years. Afterwards, we adopt four classical spatiotemporal feature learning models on DFME to perform MER experiments to objectively verify the validity of DFME dataset. In addition, we explore different solutions to the class imbalance and key-frame sequence sampling problems in dynamic MER respectively on DFME, so as to provide a valuable reference for future research. The comprehensive experimental results show that our DFME dataset can facilitate the research of automatic MER, and provide a new benchmark for MER. DFME will be published via https://mea-lab-421.github.io.

Face Anti-spoofing (FAS) is essential to secure face recognition systems from various physical attacks. However, recent research generally focuses on short-distance applications (i.e., phone unlocking) while lacking consideration of long-distance scenes (i.e., surveillance security checks). In order to promote relevant research and fill this gap in the community, we collect a large-scale Surveillance High-Fidelity Mask (SuHiFiMask) dataset captured under 40 surveillance scenes, which has 101 subjects from different age groups with 232 3D attacks (high-fidelity masks), 200 2D attacks (posters, portraits, and screens), and 2 adversarial attacks. In this scene, low image resolution and noise interference are new challenges faced in surveillance FAS. Together with the SuHiFiMask dataset, we propose a Contrastive Quality-Invariance Learning (CQIL) network to alleviate the performance degradation caused by image quality from three aspects: (1) An Image Quality Variable module (IQV) is introduced to recover image information associated with discrimination by combining the super-resolution network. (2) Using generated sample pairs to simulate quality variance distributions to help contrastive learning strategies obtain robust feature representation under quality variation. (3) A Separate Quality Network (SQN) is designed to learn discriminative features independent of image quality. Finally, a large number of experiments verify the quality of the SuHiFiMask dataset and the superiority of the proposed CQIL.

Image Virtual try-on aims at replacing the cloth on a personal image with a garment image (in-shop clothes), which has attracted increasing attention from the multimedia and computer vision communities. Prior methods successfully preserve the character of clothing images, however, occlusion remains a pernicious effect for realistic virtual try-on. In this work, we first present a comprehensive analysis of the occlusions and categorize them into two aspects: i) Inherent-Occlusion: the ghost of the former cloth still exists in the try-on image; ii) Acquired-Occlusion: the target cloth warps to the unreasonable body part. Based on the in-depth analysis, we find that the occlusions can be simulated by a novel semantically-guided mixup module, which can generate semantic-specific occluded images that work together with the try-on images to facilitate training a de-occlusion try-on (DOC-VTON) framework. Specifically, DOC-VTON first conducts a sharpened semantic parsing on the try-on person. Aided by semantics guidance and pose prior, various complexities of texture are selectively blending with human parts in a copy-and-paste manner. Then, the Generative Module (GM) is utilized to take charge of synthesizing the final try-on image and learning to de-occlusion jointly. In comparison to the state-of-the-art methods, DOC-VTON achieves better perceptual quality by reducing occlusion effects.

This work focuses on unsupervised representation learning in person re-identification (ReID). Recent self-supervised contrastive learning methods learn invariance by maximizing the representation similarity between two augmented views of a same image. However, traditional data augmentation may bring to the fore undesirable distortions on identity features, which is not always favorable in id-sensitive ReID tasks. In this paper, we propose to replace traditional data augmentation with a generative adversarial network (GAN) that is targeted to generate augmented views for contrastive learning. A 3D mesh guided person image generator is proposed to disentangle a person image into id-related and id-unrelated features. Deviating from previous GAN-based ReID methods that only work in id-unrelated space (pose and camera style), we conduct GAN-based augmentation on both id-unrelated and id-related features. We further propose specific contrastive losses to help our network learn invariance from id-unrelated and id-related augmentations. By jointly training the generative and the contrastive modules, our method achieves new state-of-the-art unsupervised person ReID performance on mainstream large-scale benchmarks.

Due to their ability to offer more comprehensive information than data from a single view, multi-view (multi-source, multi-modal, multi-perspective, etc.) data are being used more frequently in remote sensing tasks. However, as the number of views grows, the issue of data quality becomes more apparent, limiting the potential benefits of multi-view data. Although recent deep neural network (DNN) based models can learn the weight of data adaptively, a lack of research on explicitly quantifying the data quality of each view when fusing them renders these models inexplicable, performing unsatisfactorily and inflexible in downstream remote sensing tasks. To fill this gap, in this paper, evidential deep learning is introduced to the task of aerial-ground dual-view remote sensing scene classification to model the credibility of each view. Specifically, the theory of evidence is used to calculate an uncertainty value which describes the decision-making risk of each view. Based on this uncertainty, a novel decision-level fusion strategy is proposed to ensure that the view with lower risk obtains more weight, making the classification more credible. On two well-known, publicly available datasets of aerial-ground dual-view remote sensing images, the proposed approach achieves state-of-the-art results, demonstrating its effectiveness. The code and datasets of this article are available at the following address: https://github.com/gaopiaoliang/Evidential.

Most Graph Neural Networks follow the message-passing paradigm, assuming the observed structure depicts the ground-truth node relationships. However, this fundamental assumption cannot always be satisfied, as real-world graphs are always incomplete, noisy, or redundant. How to reveal the inherent graph structure in a unified way remains under-explored. We proposed PRI-GSL, a Graph Structure Learning framework guided by the Principle of Relevant Information, providing a simple and unified framework for identifying the self-organization and revealing the hidden structure. PRI-GSL learns a structure that contains the most relevant yet least redundant information quantified by von Neumann entropy and Quantum Jensen-Shannon divergence. PRI-GSL incorporates the evolution of quantum continuous walk with graph wavelets to encode node structural roles, showing in which way the nodes interplay and self-organize with the graph structure. Extensive experiments demonstrate the superior effectiveness and robustness of PRI-GSL.

Designing better deep networks and better reinforcement learning (RL) algorithms are both important for deep RL. This work focuses on the former. Previous methods build the network with several modules like CNN, LSTM and Attention. Recent methods combine the Transformer with these modules for better performance. However, it requires tedious optimization skills to train a network composed of mixed modules, making these methods inconvenient to be used in practice. In this paper, we propose to design \emph{pure Transformer-based networks} for deep RL, aiming at providing off-the-shelf backbones for both the online and offline settings. Specifically, the Transformer in Transformer (TIT) backbone is proposed, which cascades two Transformers in a very natural way: the inner one is used to process a single observation, while the outer one is responsible for processing the observation history; combining both is expected to extract spatial-temporal representations for good decision-making. Experiments show that TIT can achieve satisfactory performance in different settings, consistently.

Patients take care of what their teeth will be like after the orthodontics. Orthodontists usually describe the expectation movement based on the original smile images, which is unconvincing. The growth of deep-learning generative models change this situation. It can visualize the outcome of orthodontic treatment and help patients foresee their future teeth and facial appearance. While previous studies mainly focus on 2D or 3D virtual treatment outcome (VTO) at a profile level, the problem of simulating treatment outcome at a frontal facial image is poorly explored. In this paper, we build an efficient and accurate system for simulating virtual teeth alignment effects in a frontal facial image. Our system takes a frontal face image of a patient with visible malpositioned teeth and the patient's 3D scanned teeth model as input, and progressively generates the visual results of the patient's teeth given the specific orthodontics planning steps from the doctor (i.e., the specification of translations and rotations of individual tooth). We design a multi-modal encoder-decoder based generative model to synthesize identity-preserving frontal facial images with aligned teeth. In addition, the original image color information is used to optimize the orthodontic outcomes, making the results more natural. We conduct extensive qualitative and clinical experiments and also a pilot study to validate our method.

Future work sentences (FWS) are the particular sentences in academic papers that contain the author's description of their proposed follow-up research direction. This paper presents methods to automatically extract FWS from academic papers and classify them according to the different future directions embodied in the paper's content. FWS recognition methods will enable subsequent researchers to locate future work sentences more accurately and quickly and reduce the time and cost of acquiring the corpus. The current work on automatic identification of future work sentences is relatively small, and the existing research cannot accurately identify FWS from academic papers, and thus cannot conduct data mining on a large scale. Furthermore, there are many aspects to the content of future work, and the subdivision of the content is conducive to the analysis of specific development directions. In this paper, Nature Language Processing (NLP) is used as a case study, and FWS are extracted from academic papers and classified into different types. We manually build an annotated corpus with six different types of FWS. Then, automatic recognition and classification of FWS are implemented using machine learning models, and the performance of these models is compared based on the evaluation metrics. The results show that the Bernoulli Bayesian model has the best performance in the automatic recognition task, with the Macro F1 reaching 90.73%, and the SCIBERT model has the best performance in the automatic classification task, with the weighted average F1 reaching 72.63%. Finally, we extract keywords from FWS and gain a deep understanding of the key content described in FWS, and we also demonstrate that content determination in FWS will be reflected in the subsequent research work by measuring the similarity between future work sentences and the abstracts.