ACM MM 2024 Papers

Exposure correction aims to enhance visual data suffering from improper exposures, which can greatly improve satisfactory visual effects. However, previous methods mainly focus on the image modality, and the video counterpart is less explored in the literature. Directly applying prior image-based methods to videos results in temporal incoherence with low visual quality. Through thorough investigation, we find that the development of relevant communities is limited by the absence of a benchmark dataset. Therefore, in this paper, we construct the first real-world paired video dataset, including both underexposure and overexposure dynamic scenes. To achieve spatial alignment, we utilize two DSLR cameras and a beam splitter to simultaneously capture improper and normal exposure videos. Additionally, we propose an end-to-end video exposure correction network, in which a dual-stream module is designed to deal with both underexposure and overexposure factors, enhancing the illumination based on Retinex theory. The extensive experiments based on various metrics and user studies demonstrate the significance of our dataset and the effectiveness of our method. The code and dataset are available at https://github.com/kravrolens/VECNet.

This paper presents a pilot study aimed at introducing multi-agent debate into multimodal reasoning. The study addresses two key challenges: the trivialization of opinions resulting from excessive summarization and the diversion of focus caused by distractor concepts introduced from images. These challenges stem from the inductive (bottom-up) nature of existing debating schemes. To address the issue, we propose a deductive (top-down) debating approach called Blueprint Debate on Graphs (BDoG). In BDoG, debates are confined to a blueprint graph to prevent opinion trivialization through world-level summarization. Moreover, by storing evidence in branches within the graph, BDoG mitigates distractions caused by frequent but irrelevant concepts. Extensive experiments validate that BDoG is able to achieve state-of-the-art results in ScienceQA and MMBench with significant improvements over previous methods. The source code can be accessed at https://github.com/thecharm/BDoG.

In the information retrieval scenario, query augmentation is an essential technique to refine semantically imprecise queries to align more closely with users' actual information needs. Traditional methods typically rely on extracting signals from user interactions such as browsing or clicking behaviors to augment the queries, which may not accurately reflect the actual user intent due to inherent noise and the dependency on initial user interactions. To overcome these limitations, we introduce Brain-Aug, a novel approach that decodes semantic information directly from brain signals of users to augment query representation. Brain-Aug builds on three techniques: (i) Structurally, an adapter network is utilized to project brain signals into the embedding space of a language model, allowing query augmentation conditioned on both the users' initial query and their brain signals. (ii) During training, we use a next token prediction task for query augmentation and adopt prompt tuning to efficiently train the brain adapter. (iii)At the inference stage, a ranking-oriented decoding strategy is implemented, enabling Brain-Aug to generate augmentations that improve ranking performance. We evaluate our approach on multiple functional magnetic resonance imaging (fMRI) datasets, demonstrating that Brain-Aug not only produces semantically richer queries but also significantly improves document ranking accuracy, particularly for ambiguous queries. These results validate the effectiveness of Brain-Aug, and reveal the potential of using internal cognitive states to understand and augment text-based queries. Supplementary materials and code are available at https://github.com/YeZiyi1998/Brain-Query-Augmentation.

Depression is a prevalent mental health disorder that significantly impacts individuals' lives and well-being. Early detection and intervention are crucial for effective treatment and management of depression. Recently, there are many end-to-end deep learning methods leveraging the facial expression features for automatic depression detection. However, most current methods overlook the temporal dynamics of facial expressions. Although very recent 3DCNN methods remedy this gap, they introduce more computational cost due to the selection of CNN-based backbones and redundant facial features. To address the above limitations, by considering the timing correlation of facial expressions, we propose a novel framework called FacialPulse, which recognizes depression with high accuracy and speed. By harnessing the bidirectional nature and proficiently addressing long-term dependencies, the Facial Motion Modeling Module (FMMM) is designed in FacialPulse to fully capture temporal features. Since the proposed FMMM has parallel processing capabilities and has the gate mechanism to mitigate gradient vanishing, this module can also significantly boost the training speed. Besides, to effectively use facial landmarks to replace original images to decrease information redundancy, a Facial Landmark Calibration Module (FLCM) is designed to eliminate facial landmark errors to further improve recognition accuracy. Extensive experiments on the AVEC2014 dataset and MMDA dataset (a depression dataset) demonstrate the superiority of FacialPulse on recognition accuracy and speed, with the average MAE (Mean Absolute Error) decreased by 21% compared to baselines, and the recognition speed increased by 100% compared to state-of-the-art methods. Codes are released at https://github.com/volatileee/FacialPulse.

Virtual reality (VR) is a revolutionary method of presenting data visualizations, which brings potential possibilities for enhancing analytical activities. However, applying this method to visualize complex data flows remains largely underexplored, especially the Sankey diagrams, which have an advantageous capacity to represent trends in data flows. In this work, we explored a novel design for the immersive Sankey diagram system within VR environments, utilizing a three-dimensional visual design and several interaction techniques that leveraged VR's spatial and immersive capabilities. Through two comparative user studies, we found the effectiveness of the VR Sankey diagram system in improving task performance and engagement and reducing cognitive workload in complex data analysis. We contribute an interactive, immersive Sankey diagram system in VR environments, empirical evidence of its advantages, and design lessons for future immersive visualization tools.

This paper provides an efficient training-free painterly image harmonization (PIH) method, dubbed FreePIH, that leverages only a pre-trained diffusion model to achieve state-of-the-art harmonization results. Unlike existing methods that require either training auxiliary networks or fine-tuning a large pre-trained backbone, or both, to harmonize a foreground object with a painterly-style background image, our FreePIH tames the denoising process as a plug-in module for foreground image style transfer. Specifically, we find that the very last few steps of the denoising (i.e., generation) process strongly correspond to the stylistic information of images, and based on this, we propose to augment the latent features of both the foreground and background images with Gaussians for a direct denoising-based harmonization. To guarantee the fidelity of the harmonized image, we make use of latent features to enforce the consistency of the content and stability of the foreground objects in the latent space, and meanwhile, aligning both fore-/back-grounds with the same style. Moreover, to accommodate the generation with more structural and textural details, we further integrate text prompts to attend to the latent features, hence improving the generation quality. Quantitative and qualitative evaluations on COCO and LAION 5B datasets demonstrate that our method can surpass representative baselines by large margins.

Although the diffusion model has achieved remarkable performance in the field of image generation, its high inference delay hinders its wide application in edge devices with scarce computing resources. Therefore, many training-free sampling methods have been proposed to reduce the number of sampling steps required for diffusion models. However, they perform poorly under a very small number of sampling steps. Thanks to the emergence of knowledge distillation technology, the existing training scheme methods have achieved excellent results at very low step numbers. However, the current methods mainly focus on designing novel diffusion model sampling methods with knowledge distillation. How to transfer better diffusion knowledge from teacher models is a more valuable problem but rarely studied. Therefore, we propose Relational Diffusion Distillation (RDD), a novel distillation method tailored specifically for distilling diffusion models. Unlike existing methods that simply align teacher and student models at pixel level or feature distributions, our method introduces cross-sample relationship interaction during the distillation process and alleviates the memory constraints induced by multiple sample interactions. Our RDD significantly enhances the effectiveness of the progressive distillation framework within the diffusion model. Extensive experiments on several datasets (e.g., CIFAR-10 and ImageNet) demonstrate that our proposed RDD leads to 1.47 FID decrease under 1 sampling step compared to state-of-the-art diffusion distillation methods and achieving 256x speed-up compared to DDIM strategy. Code is available at https://github.com/cantbebetter2/RDD.

In the domain of generative multimedia and interactive experiences, generating realistic and accurate full-body poses from sparse tracking is crucial for many real-world applications, while achieving sequence modeling and efficient motion generation remains challenging. Recently, state space models (SSMs) with efficient hardware-aware designs (i.e., Mamba) have shown great potential for sequence modeling, particularly in temporal contexts. However, processing motion data is still challenging for SSMs. Specifically, the sparsity of input conditions makes motion generation an ill-posed problem. Moreover, the complex structure of the human body further complicates this task. To address these issues, we present Motion Mamba Diffusion (MMD), a novel conditional diffusion model, which effectively utilizes the sequence modeling capability of SSMs and the robust generation ability of diffusion models to track full-body poses accurately. In particular, we design a bidirectional Temporal Mamba Module (TMM) to model motion sequence. Additionally, a Spatial Mamba Module (SMM) is further proposed for feature enhancement within a single frame. Extensive experiments on the large motion capture dataset (AMASS) demonstrate that our proposed approach outperforms the latest methods in terms of accuracy and smoothness, thus providing a crucial advancement for creating realistic virtual avatars in various applications.

The neural radiance fields (NeRF) have advanced the development of 3D volumetric video technology, but the large data volumes they involve pose significant challenges for storage and transmission. To address these problems, the existing solutions typically compress these NeRF representations after the training stage, leading to a separation between representation training and compression. In this paper, we try to directly learn a compact NeRF representation for volumetric video in the training stage based on the proposed rate-aware compression framework. Specifically, for volumetric video, we use a simple yet effective modeling strategy to reduce temporal redundancy for the NeRF representation. Then, during the training phase, an implicit entropy model is utilized to estimate the bitrate of the NeRF representation. This entropy model is then encoded into the bitstream to assist in the decoding of the NeRF representation. This approach enables precise bitrate estimation, thereby leading to a compact NeRF representation.Furthermore, we propose an adaptive quantization strategy and learn the optimal quantization step for the NeRF representations. Finally, the NeRF representation can be optimized by using the rate-distortion trade-off. Our proposed compression framework can be used for different representations and experimental results demonstrate that our approach significantly reduces the storage size with marginal distortion and achieves state-of-the-art rate-distortion performance for volumetric video on the HumanRF and ReRF datasets. Compared to the previous state-of-the-art method TeTriRF, we achieved an approximately -80% BD-rate on the HumanRF dataset and -60% BD-rate on the ReRF dataset.

While existing Aspect-based Sentiment Analysis (ABSA) has received extensive effort and advancement, there are still gaps in defining a more holistic research target seamlessly integrating multimodality, conversation context, fine-granularity, and also covering the changing sentiment dynamics as well as cognitive causal rationales. This paper bridges the gaps by introducing a multimodal conversational ABSA, where two novel subtasks are proposed: 1) Panoptic Sentiment Sextuple Extraction, panoramically recognizing holder, target, aspect, opinion, sentiment, rationale from multi-turn multi-party multimodal dialogue. 2) Sentiment Flipping Analysis, detecting the dynamic sentiment transformation throughout the conversation with the causal reasons. To benchmark the tasks, we construct PanoSent, a dataset annotated both manually and automatically, featuring high quality, large scale, multimodality, multilingualism, multi-scenarios, and covering both implicit&explicit sentiment elements. To effectively address the tasks, we devise a novel Chain-of-Sentiment reasoning framework, together with a novel multimodal large language model (namely Sentica) and a paraphrase-based verification mechanism. Extensive evaluations demonstrate the superiority of our methods over strong baselines, validating the efficacy of all our proposed methods. The work is expected to open up a new era for the ABSA community, and thus all our codes and data are open at https://PanoSent.github.io/.

Reconstructing dynamic scenes from video sequences is a highly promising task in the multimedia domain. While previous methods have made progress, they often struggle with slow rendering and managing temporal complexities such as significant motion and object appearance/disappearance. In this paper, we propose SaRO-GS as a novel dynamic scene representation capable of achieving real-time rendering while effectively handling temporal complexities in dynamic scenes. To address the issue of slow rendering speed, we adopt a Gaussian primitive-based representation and optimize the Gaussians in 4D space, which facilitates real-time rendering with the assistance of 3D Gaussian Splatting. Additionally, to handle temporally complex dynamic scenes, we introduce a Scale-aware Residual Field. This field considers the size information of each Gaussian primitive while encoding its residual feature and aligns with the self-splitting behavior of Gaussian primitives. Furthermore, we propose an Adaptive Optimization Schedule, which assigns different optimization strategies to Gaussian primitives based on their distinct temporal properties, thereby expediting the reconstruction of dynamic regions. Through evaluations on monocular and multi-view datasets, our method has demonstrated state-of-the-art performance. Please see our project page at https://yjb6.github.io/SaRO-GS.github.io.

Humans exhibit a remarkable ability to learn quickly from a limited number of labeled samples, a capability that starkly contrasts with that of current machine learning systems. Unsupervised Few-Shot Learning (U-FSL) seeks to bridge this divide by reducing reliance on annotated datasets during initial training phases. In this work, we first quantitatively assess the impacts of Masked Image Modeling (MIM) and Contrastive Learning (CL) on few-shot learning tasks. Our findings highlight the respective limitations of MIM and CL in terms of discriminative and generalization abilities, which contribute to their underperformance in U-FSL contexts. To address these trade-offs between generalization and discriminability in unsupervised pretraining, we introduce a novel paradigm named Masked Image Contrastive Modeling (MICM). MICM creatively combines the targeted object learning strength of CL with the generalized visual feature learning capability of MIM, significantly enhancing its efficacy in downstream few-shot learning inference. Extensive experimental analyses confirm the advantages of MICM, demonstrating significant improvements in both generalization and discrimination capabilities for few-shot learning. Our comprehensive quantitative evaluations further substantiate the superiority of MICM, showing that our two-stage U-FSL framework based on MICM markedly outperforms existing leading baselines. The repository of this project is available at https://github.com/iCGY96/MICM.

Tracking by detection has been the prevailing paradigm in the field of Multi-object Tracking (MOT). These methods typically rely on the Kalman Filter to estimate the future locations of objects, assuming linear object motion. However, they fall short when tracking objects exhibiting nonlinear and diverse motion in scenarios like dancing and sports. In addition, there has been limited focus on utilizing learning-based motion predictors in MOT. To address these challenges, we resort to exploring data-driven motion prediction methods. Inspired by the great expectation of state space models (SSMs), such as Mamba, in long-term sequence modeling with near-linear complexity, we introduce a Mamba-based motion model named Mamba moTion Predictor (MTP). MTP is designed to model the complex motion patterns of objects like dancers and athletes. Specifically, MTP takes the spatial-temporal location dynamics of objects as input, captures the motion pattern using a bi-Mamba encoding layer, and predicts the next motion. In real-world scenarios, objects may be missed due to occlusion or motion blur, leading to premature termination of their trajectories. To tackle this challenge, we further expand the application of MTP. We employ it in an autoregressive way to compensate for missing observations by utilizing its own predictions as inputs, thereby contributing to more consistent trajectories. Our proposed tracker, MambaTrack, demonstrates advanced performance on benchmarks such as Dancetrack and SportsMOT, which are characterized by complex motion and severe occlusion.

Recent advancements in "deepfake" techniques have paved the way for generating various media forgeries. In response to the potential hazards of these media forgeries, many researchers engage in exploring detection methods, increasing the demand for high-quality media forgery datasets. Despite this, existing datasets have certain limitations. Firstly, most datasets focus on manipulating visual modality and usually lack diversity, as only a few forgery approaches are considered. Secondly, the quality of media is often inadequate in clarity and naturalness. Meanwhile, the size of the dataset is also limited. Thirdly, it is commonly observed that real-world forgeries are motivated by identity, yet the identity information of the individuals portrayed in these forgeries within existing datasets remains under-explored. For detection, identity information could be an essential clue to boost performance. Moreover, official media concerning relevant identities on the Internet can serve as prior knowledge, aiding both the audience and forgery detectors in determining the true identity. Therefore, we propose an identity-driven multimedia forgery dataset, IDForge, which contains 249,138 video shots sourced from 324 wild videos of 54 celebrities collected from the Internet. The fake video shots involve 9 types of manipulation across visual, audio, and textual modalities. Additionally, IDForge provides extra 214,438 real video shots as a reference set for the 54 celebrities. Correspondingly, we propose the Reference-assisted Multimodal Forgery Detection Network (R-MFDN), aiming at the detection of deepfake videos. Through extensive experiments on the proposed dataset, we demonstrate the effectiveness of R-MFDN on the multimedia detection task. The dataset is available at: https://github.com/xyyandxyy/IDForge.

Classifying videos differs from that of images in the need to capture the information on what has happened, instead of what is in the frames. Conventional methods typically follow the data-driven approach, which uses transformer-based attention models to extract and aggregate the features of video frames as the representation of the entire video. However, this approach tends to extract the object information of frames and may face difficulties in classifying the classes talking about events, such as "fixing bicycle". To address this issue, This paper presents an Event-level Causal Representation Learning (ECRL) model for the spatio-temporal modeling of both the in-frame object interactions and their cross-frame temporal correlations. Specifically, ECRL first employs a Frame-to-Video Causal Modeling (F2VCM) module, which simultaneously builds the in-frame causal graph with the background and foreground information and models their cross-frame correlations to construct a video-level causal graph. Subsequently, a Causality-aware Event-level Representation Inference (CERI) module is introduced to eliminate the spurious correlations in contexts and objects via the back- and front-door interventions, respectively. The former involves visual context de-biasing to filter out background confounders, while the latter employs global-local causal attention to capture event-level visual information. Experimental results on two benchmarking datasets verified that ECRL may better capture the cross-frame correlations to describe videos in event-level features. The source codes have been released at https://github.com/wyqcrystal/ECRL.

Diffusion models have shown impressive potential on talking head generation. While plausible appearance and talking effect are achieved, these methods still suffer from temporal, 3D or expression inconsistency due to the error accumulation and inherent limitation of single-image generation ability. In this paper, we propose ConsistentAvatar, a novel framework for fully consistent and high-fidelity talking avatar generation. Instead of directly employing multi-modal conditions to the diffusion process, our method learns to first model the temporal representation for stability between adjacent frames. Specifically, we propose a Temporally-Sensitive Detail (TSD) map containing high-frequency feature and contours that vary significantly along the time axis. Using a temporal consistent diffusion module, we learn to align TSD of the initial result to that of the video frame ground truth. The final avatar is generated by a fully consistent diffusion module, conditioned on the aligned TSD, rough head normal, and emotion prompt embedding. We find that the aligned TSD, which represents the temporal patterns, constrains the diffusion process to generate temporally stable talking head. Further, its reliable guidance complements the inaccuracy of other conditions, suppressing the accumulated error while improving the consistency on various aspects. Extensive experiments demonstrate that ConsistentAvatar outperforms the state-of-the-art methods on the generated appearance, 3D, expression and temporal consistency.

The detection and localization of highly realistic deepfake audio-visual content are challenging even for the most advanced state-of-the-art methods. While most of the research efforts in this domain are focused on detecting high-quality deepfake images and videos, only a few works address the problem of the localization of small segments of audio-visual manipulations embedded in real videos. In this research, we emulate the process of such content generation and propose the AV-Deepfake1M dataset. The dataset contains content-driven (i) video manipulations, (ii) audio manipulations, and (iii) audio-visual manipulations for more than 2K subjects resulting in a total of more than 1M videos. The paper provides a thorough description of the proposed data generation pipeline accompanied by a rigorous analysis of the quality of the generated data. The comprehensive benchmark of the proposed dataset utilizing state-of-the-art deepfake detection and localization methods indicates a significant drop in performance compared to previous datasets. The proposed dataset will play a vital role in building the next-generation deepfake localization methods. The dataset and associated code are available at https://github.com/ControlNet/AV-Deepfake1M.

In the Fourier frequency domain, luminance information is primarily encoded in the amplitude component, while spatial structure information is significantly contained within the phase component. Existing low-light image enhancement techniques using Fourier transform have mainly focused on amplifying the amplitude component and simply replicating the phase component, an approach that often leads to color distortions and noise issues. In this paper, we propose a Dual-Stage Multi-Branch Fourier Low-Light Image Enhancement (DMFourLLIE) framework to address these limitations by emphasizing the phase component's role in preserving image structure and detail. The first stage integrates structural information from infrared images to enhance the phase component and employs a luminance-attention mechanism in the luminance-chrominance color space to precisely control amplitude enhancement. The second stage combines multi-scale and Fourier convolutional branches for robust image reconstruction, effectively recovering spatial structures and textures. This dual-branch joint optimization process ensures that complex image information is retained, overcoming the limitations of previous methods that neglected the interplay between amplitude and phase. Extensive experiments across multiple datasets demonstrate that DMFourLLIE outperforms current state-of-the-art methods in low-light image enhancement.