Despite the impressive progress of multimodal generative models, video-to-audio generation still suffers from limited performance and limits the flexibility to prioritize sound synthesis for specific objects within the scene. Conversely, text-to-audio generation methods generate high-quality audio but pose challenges in ensuring comprehensive scene depiction and time-varying control. To tackle these challenges, we propose a novel video-and-text-to-audio generation method, called ReWaS, where video serves as a conditional control for a text-to-audio generation model. Especially, our method estimates the structural information of sound (namely, energy) from the video while receiving key content cues from a user prompt. We employ a well-performing text-to-audio model to consolidate the video control, which is much more efficient for training multimodal diffusion models with massive triplet-paired (audio-video-text) data. In addition, by separating the generative components of audio, it becomes a more flexible system that allows users to freely adjust the energy, surrounding environment, and primary sound source according to their preferences. Experimental results demonstrate that our method shows superiority in terms of quality, controllability, and training efficiency.

View-invariant representation learning from egocentric (first-person, ego) and exocentric (third-person, exo) videos is a promising approach toward generalizing video understanding systems across multiple viewpoints. However, this area has been underexplored due to the substantial differences in perspective, motion patterns, and context between ego and exo views. In this paper, we propose a novel masked ego-exo modeling that promotes both causal temporal dynamics and cross-view alignment, called Bootstrap Your Own Views (BYOV), for fine-grained view-invariant video representation learning from unpaired ego-exo videos. We highlight the importance of capturing the compositional nature of human actions as a basis for robust cross-view understanding. Specifically, self-view masking and cross-view masking predictions are designed to learn view-invariant and powerful representations concurrently. Experimental results demonstrate that our BYOV significantly surpasses existing approaches with notable gains across all metrics in four downstream ego-exo video tasks. The code is available at https://github.com/park-jungin/byov.

Person search aims to localize a person of interest in a large image gallery captured by multiple, non-overlapping cameras. Prevalent unified methods have suffered from (1) noisy proposals with mis-detection and occlusion, and (2) large appearance variation within a class, which deteriorates the prototype-based metric learning. To address these problems, we introduce a Prototype-guided Attention Distillation, shortly PAD, which exploits a prototype (a typical representation of an identity) as a guidance to the attention module to consistently highlight identity-inherent regions across different poses. To utilize the knowledge encoded in prototypes for matching unseen IDs, PAD conducts attention distillation to guide student Re-ID queries by deeply mimicking attention maps from the prototype query. Additionally, to address large intra-class variation induced by pose or camera views, we extend PAD with multiple part prototypes representing consistent local regions across different instances. Furthermore, we exploit an adaptive momentum strategy for robust attention distillation in PAD to update more distinct prototypes. Extensive experiments conducted on CUHK-SYSU and PRW demonstrate the effectiveness of PAD, showcasing state-of-the-art performance. Moreover, our distilled attention surprisingly highlights distinguished multiple regions for person search.

This paper introduces VLAP, a novel approach that bridges pretrained vision models and large language models (LLMs) to make frozen LLMs understand the visual world. VLAP transforms the embedding space of pretrained vision models into the LLMs' word embedding space using a single linear layer for efficient and general-purpose visual and language understanding. Specifically, we harness well-established word embeddings to bridge two modality embedding spaces. The visual and text representations are simultaneously assigned to a set of word embeddings within pretrained LLMs by formulating the assigning procedure as an optimal transport problem. We predict the assignment of one modality from the representation of another modality data, enforcing consistent assignments for paired multimodal data. This allows vision and language representations to contain the same information, grounding the frozen LLMs' word embedding space in visual data. Moreover, a robust semantic taxonomy of LLMs can be preserved with visual data since the LLMs interpret and reason linguistic information from correlations between word embeddings. Experimental results show that VLAP achieves substantial improvements over the previous linear transformation-based approaches across a range of vision-language tasks, including image captioning, visual question answering, and cross-modal retrieval. We also demonstrate the learned visual representations hold a semantic taxonomy of LLMs, making visual semantic arithmetic possible.

Text-to-3D generation has shown rapid progress in recent days with the advent of score distillation sampling (SDS), a methodology of using pretrained text-to-2D diffusion models to optimize a neural radiance field (NeRF) in a zero-shot setting. However, the lack of 3D awareness in the 2D diffusion model often destabilizes previous methods from generating a plausible 3D scene. To address this issue, we propose 3DFuse, a novel framework that incorporates 3D awareness into the pretrained 2D diffusion model, enhancing the robustness and 3D consistency of score distillation-based methods. Specifically, we introduce a consistency injection module which constructs a 3D point cloud from the text prompt and utilizes its projected depth map at given view as a condition for the diffusion model. The 2D diffusion model, through its generative capability, robustly infers dense structure from the sparse point cloud depth map and generates a geometrically consistent and coherent 3D scene. We also introduce a new technique called semantic coding that reduces semantic ambiguity of the text prompt for improved results. Our method can be easily adapted to various text-to-3D baselines, and we experimentally demonstrate how our method notably enhances the 3D consistency of generated scenes in comparison to previous baselines, achieving state-of-the-art performance in geometric robustness and fidelity.

We present a novel method for exemplar-based image translation, called matching interleaved diffusion models (MIDMs). Most existing methods for this task were formulated as GAN-based matching-then-generation framework. However, in this framework, matching errors induced by the difficulty of semantic matching across cross-domain, e.g., sketch and photo, can be easily propagated to the generation step, which in turn leads to the degenerated results. Motivated by the recent success of diffusion models, overcoming the shortcomings of GANs, we incorporate the diffusion models to overcome these limitations. Specifically, we formulate a diffusion-based matching-and-generation framework that interleaves cross-domain matching and diffusion steps in the latent space by iteratively feeding the intermediate warp into the noising process and denoising it to generate a translated image. In addition, to improve the reliability of diffusion process, we design confidence-aware process using cycle-consistency to consider only confident regions during translation. Experimental results show that our MIDMs generate more plausible images than state-of-the-art methods.

In this paper, we efficiently transfer the surpassing representation power of the vision foundation models, such as ViT and Swin, for video understanding with only a few trainable parameters. Previous adaptation methods have simultaneously considered spatial and temporal modeling with a unified learnable module but still suffered from fully leveraging the representative capabilities of image transformers. We argue that the popular dual-path (two-stream) architecture in video models can mitigate this problem. We propose a novel DualPath adaptation separated into spatial and temporal adaptation paths, where a lightweight bottleneck adapter is employed in each transformer block. Especially for temporal dynamic modeling, we incorporate consecutive frames into a grid-like frameset to precisely imitate vision transformers' capability that extrapolates relationships between tokens. In addition, we extensively investigate the multiple baselines from a unified perspective in video understanding and compare them with DualPath. Experimental results on four action recognition benchmarks prove that pretrained image transformers with DualPath can be effectively generalized beyond the data domain.

The goal of this work is zero-shot text-to-speech synthesis, with speaking styles and voices learnt from facial characteristics. Inspired by the natural fact that people can imagine the voice of someone when they look at his or her face, we introduce a face-styled diffusion text-to-speech (TTS) model within a unified framework learnt from visible attributes, called Face-TTS. This is the first time that face images are used as a condition to train a TTS model.We jointly train cross-model biometrics and TTS models to preserve speaker identity between face images and generated speech segments. We also propose a speaker feature binding loss to enforce the similarity of the generated and the ground truth speech segments in speaker embedding space. Since the biometric information is extracted directly from the face image, our method does not require extra fine-tuning steps to generate speech from unseen and unheard speakers. We train and evaluate the model on the LRS3 dataset, an in-the-wild audio-visual corpus containing background noise and diverse speaking styles. The project page is https://facetts.github.io.

Existing text-to-image diffusion models struggle to synthesize realistic images given dense captions, where each text prompt provides a detailed description for a specific image region. To address this, we propose DenseDiffusion, a training-free method that adapts a pre-trained text-to-image model to handle such dense captions while offering control over the scene layout. We first analyze the relationship between generated images’ layouts and the pre-trained model’s intermediate attention maps. Next, we develop an attention modulation method that guides objects to appear in specific regions according to layout guidance. Without requiring additional fine-tuning or datasets, we improve image generation performance given dense captions regarding both automatic and human evaluation scores. In addition, we achieve similar-quality visual results with models specifically trained with layout conditions. Code and data are available at https://github.com/naver-ai/DenseDiffusion.

Compositional zero-shot learning (CZSL) aims to recognize unseen compositions with prior knowledge of known primitives (attribute and object). Previous works for CZSL often suffer from grasping the contextuality between attribute and object, as well as the discriminability of visual features, and the long-tailed distribution of real-world compositional data. We propose a simple and scalable framework called Composition Transformer (CoT) to address these issues. CoT employs object and attribute experts in distinctive manners to generate representative embeddings, using the visual network hierarchically. The object expert extracts representative object embeddings from the final layer in a bottom-up manner, while the attribute expert makes attribute embeddings in a top-down manner with a proposed object-guided attention module that models contextuality explicitly. To remedy biased prediction caused by imbalanced data distribution, we develop a simple minority attribute augmentation (MAA) that synthesizes virtual samples by mixing two images and oversampling minority attribute classes. Our method achieves SoTA performance on several benchmarks, including MIT-States, C-GQA, and VAW-CZSL. We also demonstrate the effectiveness of CoT in improving visual discrimination and addressing the model bias from the imbalanced data distribution. The code is available at https://github.com/HanjaeKim98/CoT.

Multi-resolution hash encoding has recently been proposed to reduce the computational cost of neural renderings, such as NeRF. This method requires accurate camera poses for the neural renderings of given scenes. However, contrary to previous methods jointly optimizing camera poses and 3D scenes, the naive gradient-based camera pose refinement method using multi-resolution hash encoding severely deteriorates performance. We propose a joint optimization algorithm to calibrate the camera pose and learn a geometric representation using efficient multi-resolution hash encoding. Showing that the oscillating gradient flows of hash encoding interfere with the registration of camera poses, our method addresses the issue by utilizing smooth interpolation weighting to stabilize the gradient oscillation for the ray samplings across hash grids. Moreover, the curriculum training procedure helps to learn the level-wise hash encoding, further increasing the pose refinement. Experiments on the novel-view synthesis datasets validate that our learning frameworks achieve state-of-the-art performance and rapid convergence of neural rendering.

Given an untrimmed video and a language query depicting a specific temporal moment in the video, video grounding aims to localize the time interval by understanding the text and video simultaneously. One of the most challenging issues is an extremely time- and cost-consuming annotation collection, including video captions in a natural language form and their corresponding temporal regions. In this paper, we present a simple yet novel training framework for video grounding in the zero-shot setting, which learns a network with only video data without any annotation. Inspired by the recent language-free paradigm, i.e. training without language data, we train the network without compelling the generation of fake (pseudo) text queries into a natural language form. Specifically, we propose a method for learning a video grounding model by selecting a temporal interval as a hypothetical correct answer and considering the visual feature selected by our method in the interval as a language feature, with the help of the well-aligned visual-language space of CLIP. Extensive experiments demonstrate the prominence of our language-free training framework, outperforming the existing zero-shot video grounding method and even several weakly-supervised approaches with large margins on two standard datasets.

The rise of deep neural networks has led to several break-throughs for semantic segmentation. In spite of this, a model trained on source domain often fails to work properly in new challenging domains, that is directly concerned with the generalization capability of the model. In this paper, we present a novel memory-guided domain generalization method for semantic segmentation based on meta-learning framework. Especially, our method abstracts the conceptual knowledge of semantic classes into categorical memory which is constant beyond the domains. Upon the meta-learning concept, we repeatedly train memory-guided networks and simulate virtual test to 1) learn how to memorize a domain-agnostic and distinct information of classes and 2) offer an externally settled memory as a class-guidance to reduce the ambiguity of representation in the test data of arbitrary unseen domain. To this end, we also propose memory divergence and feature cohesion losses, which encourage to learn memory reading and update processes for category-aware domain generalization. Extensive experiments for semantic segmentation demonstrate the superior generalization capability of our method over state-of-the-art works on various benchmarks.11https://github.com/Genie-Kim/PintheMemory

This paper presents Probabilistic Video Contrastive Learning, a self-supervised representation learning method that bridges contrastive learning with probabilistic representation. We hypothesize that the clips composing the video have different distributions in short-term duration, but can represent the complicated and sophisticated video distribution through combination in a common embedding space. Thus, the proposed method represents video clips as normal distributions and combines them into a Mixture of Gaussians to model the whole video distribution. By sampling embeddings from the whole video distribution, we can circumvent the careful sampling strategy or transformations to generate augmented views of the clips, unlike previous deterministic methods that have mainly focused on such sample generation strategies for contrastive learning. We further propose a stochastic contrastive loss to learn proper video distributions and handle the inherent uncertainty from the nature of the raw video. Experimental results verify that our probabilistic embedding stands as a state-of-the-art video representation learning for action recognition and video retrieval on the most popular benchmarks, including UCF101 and HMDB51.

Over the past few years, image-to-image (I2I) translation methods have been proposed to translate a given image into diverse outputs. Despite the impressive results, they mainly focus on the I2I translation between two domains, so the multi-domain I2I translation still remains a challenge. To address this problem, we propose a novel multi-domain unsupervised image-to-image translation (MDUIT) framework that leverages the decomposed content feature and appearance adaptive convolution to translate an image into a target appearance while preserving the given geometric content. We also exploit a contrast learning objective, which improves the disentanglement ability and effectively utilizes multi-domain image data in the training process by pairing the semantically similar images. This allows our method to learn the diverse mappings between multiple visual domains with only a single framework. We show that the proposed method produces visually diverse and plausible results in multiple domains compared to the state-of-the-art methods.

Text-to-image generation and image captioning are recently emerged as a new experimental paradigm to assess machine intelligence. They predict continuous quantity accompanied by their sampling techniques in the generation, making evaluation complicated and intractable to get marginal distributions. Based on a recent trend that multimodal generative evaluations exploit a vison-and-language pre-trained model, we propose the negative Gaussian cross-mutual information using the CLIP features as a unified metric, coined by Mutual Information Divergence (MID). To validate, we extensively compare it with competing metrics using carefully-generated or human-annotated judgments in text-to-image generation and image captioning tasks. The proposed MID significantly outperforms the competitive methods by having consistency across benchmarks, sample parsimony, and robustness toward the exploited CLIP model. We look forward to seeing the underrepresented implications of the Gaussian cross-mutual information in multimodal representation learning and future works based on this novel proposition.

In this paper, we address the problem of separating individual speech signals from videos using audio-visual neural processing. Most conventional approaches utilize frame-wise matching criteria to extract shared information between co-occurring audio and video. Thus, their performance heavily depends on the accuracy of audio-visual synchronization and the effectiveness of their representations. To overcome the frame discontinuity problem between two modalities due to transmission delay mismatch or jitter, we propose a cross-modal affinity network (CaffNet) that learns global correspondence as well as locally-varying affinities between audio and visual streams. Given that the global term provides stability over a temporal sequence at the utterance-level, this resolves the label permutation problem characterized by inconsistent assignments. By extending the proposed cross-modal affinity on the complex network, we further improve the separation performance in the complex spectral domain. Experimental results verify that the proposed methods outperform conventional ones on various datasets, demonstrating their advantages in real-world scenarios.

This paper presents a novel method, termed Bridge to Answer, to infer correct answers for questions about a given video by leveraging adequate graph interactions of heterogeneous crossmodal graphs. To realize this, we learn question conditioned visual graphs by exploiting the relation between video and question to enable each visual node using question-to-visual interactions to encompass both visual and linguistic cues. In addition, we propose bridged visual-to-visual interactions to incorporate two complementary visual information on appearance and motion by placing the question graph as an intermediate bridge. This bridged architecture allows reliable message passing through compositional semantics of the question to generate an appropriate answer. As a result, our method can learn the question conditioned visual representations attributed to appearance and motion that show powerful capability for video question answering. Extensive experiments prove that the proposed method provides effective and superior performance than state-of-the-art methods on several benchmarks.

Domain generalization aims to learn a prediction model on multi-domain source data such that the model can generalize to a target domain with unknown statistics. Most existing approaches have been developed under the assumption that the source data is well-balanced in terms of both domain and class. However, real-world training data collected with different composition biases often exhibits severe distribution gaps for domain and class, leading to substantial performance degradation. In this paper, we propose a self-balanced domain generalization framework that adaptively learns the weights of losses to alleviate the bias caused by different distributions of the multi-domain source data. The self-balanced scheme is based on an auxiliary reweighting network that iteratively updates the weight of loss conditioned on the domain and class information by leveraging balanced meta data. Experimental results demonstrate the effectiveness of our method overwhelming state-of-the-art works for domain generalization.

Recent deep neural networks based methods have achieved state-of-the-art performance on various facial expression recognition tasks. Despite such progress, previous researches for facial expression recognition have mainly focused on analyzing color recording videos only. However, the complex emotions expressed by people with different skin colors under different lighting conditions through dynamic facial expressions can be fully understandable by integrating information from multi-modal videos. We present a novel method to estimate dimensional emotion states, where color, depth, and thermal recording videos are used as a multi-modal input. Our networks, called multi-modal recurrent attention networks (MRAN), learn spatiotemporal attention volumes to robustly recognize the facial expression based on attention-boosted feature volumes. We leverage the depth and thermal sequences as guidance priors for color sequence to selectively focus on emotional discriminative regions. We also introduce a novel benchmark for multi-modal facial expression recognition, termed as multi-modal arousal-valence facial expression recognition (MAVFER), which consists of color, depth, and thermal recording videos with corresponding continuous arousal-valence scores. The experimental results show that our method can achieve the state-of-the-art results in dimensional facial expression recognition on color recording datasets including RECOLA, SEWA and AFEW, and a multi-modal recording dataset including MAVFER.

Traditional techniques for emotion recognition have focused on the facial expression analysis only, thus providing limited ability to encode context that comprehensively represents the emotional responses. We present deep networks for context-aware emotion recognition, called CAER-Net, that exploit not only human facial expression but also context information in a joint and boosting manner. The key idea is to hide human faces in a visual scene and seek other contexts based on an attention mechanism. Our networks consist of two sub-networks, including two-stream encoding networks to separately extract the features of face and context regions, and adaptive fusion networks to fuse such features in an adaptive fashion. We also introduce a novel benchmark for context-aware emotion recognition, called CAER, that is appropriate than existing benchmarks both qualitatively and quantitatively. On several benchmarks, CAER-Net proves the effect of context for emotion recognition. Our dataset is available at http://caer-dataset.github.io.

This paper presents a novel deep architecture for weakly-supervised temporal action localization that not only generates segment-level action responses but also propagates segment-level responses to the neighborhood in a form of graph Laplacian regularization. Specifically, our approach consists of two sub-modules; a class activation module to estimate the action score map over time through the action classifiers, and a graph regularization module to refine the estimated action score map by solving a quadratic programming problem with the predicted segment-level semantic affinities. Since these two modules are integrated with fully differentiable layers, the proposed networks can be jointly trained in an end-to-end manner. Experimental results on Thumos14 and ActivityNet1.2 demonstrate that the proposed method provides outstanding performances in weakly-supervised temporal action localization.

We propose a spatiotemporal attention based deep neural networks for dimensional emotion recognition in facial videos. To learn the spatiotemporal attention that selectively focuses on emotional sailient parts within facial videos, we formulate the spatiotemporal encoder-decoder network using Convolutional LSTM (ConvLSTM) modules, which can be learned implicitly without any pixel-level annotations. By leveraging the spatiotemporal attention, we also formulate the 3D convolutional neural networks (3D-CNNs) to robustly recognize the dimensional emotion in facial videos. The experimental results show that our method can achieve the state-of-the-art results in dimensional emotion recognition with the highest concordance correlation coefficient (CCC) on RECOLA and AV+EC 2017 dataset.

We present a multi-scale deep convolutional neural network (CNN) for the task of automatic 2D-to-3D conversion. Traditional methods, which make a virtual view from a reference view, consist of separate stages i.e., depth (or disparity) estimation for the reference image and depth image-based rendering (DIBR) with estimated depth. In contrast, we reformulate the view synthesis task as an image reconstruction problem with a spatial transformer module and directly make stereo image pairs with a unified CNN framework without ground-truth depth as a supervision. We further propose a multi-scale deep architecture to capture the large displacements between images from coarse-level and enhance the detail from fine-level. Experimental results demonstrate the effectiveness of the proposed method over state-of-the-art approaches both qualitatively and quantitatively on the KITTI driving dataset.