Classifying anatomical regions in endoscopic ENT (ear, nose, and throat) images is challenging due to strong inter-class similarities, bilateral symmetry, and the scarcity of annotated datasets. To overcome these issues, we present HyMoENet. This novel hybrid deep learning architecture combines convolutional neural networks (CNNs) for localized feature extraction with Vision Transformers to represent global context. Furthermore, it leverages a sparse Mixture-of-Experts (MoE) technique to improve multi-perspective specialization. Our architecture makes use of parallel CNN-Transformer encoders, which are incorporated into a dynamic MoE layer that adaptively routes representations to the most appropriate experts. Concurrently, a semantic-preserving skip connection preserves global coherence. When tested on a clinically annotated ENT endoscopy dataset from Thong Nhat Hospital in Vietnam, HyMoENet outperformed both single-stream and conventional hybrid models with an accuracy of 97.50. These results demonstrate that integrating local-global representation learning and expert modularization enhances classification accuracy for anatomically similar structures. HyMoENet sets a new benchmark for automated ENT image processing, setting the groundwork for intelligent diagnostic systems in clinical endoscopy.
论文检索
输入标题、作者或关键词，从 7,876 篇学术成果中精准定位
Automated analysis of endoscopic imagery is a critical yet underdeveloped component of ENT (ear, nose, and throat) care, hindered by variability in devices and operators, subtle and localized findings, and fine-grained distinctions such as laterality and vocal-fold state. In addition to classification, clinicians require reliable retrieval of similar cases, both visually and through concise textual descriptions. These capabilities are rarely supported by existing public benchmarks. To this end, we introduce ENTRep, the ACM Multimedia 2025 Grand Challenge on ENT endoscopy analysis, which integrates fine-grained anatomical classification with image-to-image and text-to-image retrieval under bilingual (Vietnamese and English) clinical supervision. Specifically, the dataset comprises expert-annotated images, labeled for anatomical region and normal or abnormal status, and accompanied by dual-language narrative descriptions. In addition, we define three benchmark tasks, standardize the submission protocol, and evaluate performance on public and private test splits using server-side scoring. Moreover, we report results from the top-performing teams and provide an insightful discussion.
Errors often occur during human-robot interactions, such as failing to respond, interrupting users, or providing answers that do not meet user expectations. Detecting these issues on time is crucial for making human-robot communication more natural and user-friendly. In this technical report, we present the approach proposed by our team, CFM-HRI, for the ERR@HRI 2.0 Challenge 2025, targeting the task of interaction error detection. Specifically, we propose a lightweight and efficient time-series classification approach, empowered by cross-modal alignment, to detect interaction errors more accurately and promptly. To mitigate temporal misalignment across modalities, we adopt an upsampling alignment strategy, followed by feature fusion to obtain unified representations. A sliding-window voting mechanism is then introduced to construct training samples along with corresponding ground truth labels. Several machine learning models are employed to detect errors based on the fused features. Experimental results demonstrate the effectiveness of our approach in capturing cross-modal inconsistencies and improving detection accuracy. Our approach won first place in Sub-Challenge 2 and second place in Sub-Challenge 1 of the ERR@HRI 2.0 Challenge, held in conjunction with ACM MM 2025. We provide detailed descriptions of our data processing and experimental setup, along with an analysis of the limitations of our approach and potential directions for future work. Our code is available on https://github.com/setsaile/CFM-HRI-ERR-HRI2.0.
Reliable detection of conversational errors and user-initiated corrections is critical for effective human-robot interaction (HRI). In this study, we present a comprehensive multimodal approach leveraging temporal window processing, targeted feature engineering, and a MiniRocket + Ridge classification pipeline to address the challenges introduced by the ERR@HRI 2.0 dataset. Our methodology systematically integrates multimodal data streams, including facial expressions, acoustic features, and linguistic embeddings, to predict robot failures and user reactions. Experimental results demonstrate significant improvements over baseline models in event-level detection performance. Notably, linguistic features derived from transcript embeddings emerged as the most informative modality, substantially enhancing model performance. However, we observed challenges associated with managing false positives at the event level, suggesting avenues for future refinement in adaptive thresholding and sequential post-processing techniques. Our findings underscore the importance of careful feature selection and robust temporal modelling in developing effective real-time error detection systems for conversational robots. Our code is available online. https://github.com/Ruddy202/err-hri-2.0-armas.git.
In recent years, Multimodal Sentiment Analysis (MSA) has attracted growing attention for its ability to interpret human emotions by integrating information across multiple modalities. Multimodal Conversational Aspect-based Sentiment Analysis (MCABSA) extends this research frontier by incorporating multi-party conversational contexts and requiring comprehensive extraction of sentiment elements. MCABSA presents substantial challenges, including the need to understand complex conversational contexts, integrate heterogeneous multimodal signals, and identify causal reasoning at the cognitive level. To address these challenges, we propose a two-stage Full Fine-tuning and LLM Post-processing (FLP) framework. In the first stage, we develop a multimodal caption-enhanced full fine-tuning pipeline that performs structured extraction of sextuples and sentiment flip tuples. The second stage introduces paraphrase-based sextuple verification to identify and filter low-quality sextuples for Panoptic Sentiment Sextuple Extraction (Task-1), while implementing trigger classification with a distribution alignment mechanism to determine trigger types for sentiment flipping and enhance output consistency for Sentiment Flipping Analysis (Task-2). Comprehensive experiments on both MCABSA challenge subtasks demonstrate the effectiveness of our approach, achieving 1st place on Task-1 and 3rd place on Task-2.
This paper presents our solution for the Micro-Expression Visual Question Answering (ME-VQA) task in the 2025 Facial Micro-Expression Grand Challenge (MEGC). To address the limitations of traditional micro-expression recognition (MER) methods in dynamic modeling, semantic interpretation, and natural language interaction, we propose Emotion-Qwen-VL, a fully fine-tuned multimodal large language model tailored for micro-expression understanding. Specifically, we construct a structured, instruction-based QA dataset that reformulates emotion categories and action unit (AU) annotations into natural language QA pairs, covering classification, AU detection, and causal reasoning. We then adopt a full-parameter fine-tuning strategy to guide Qwen2.5-VL in learning fine-grained temporal facial dynamics and their emotional semantics. Experimental results on the MEGC 2025 test set demonstrate that Emotion-Qwen-VL outperforms strong baselines such as Qwen2.5-VL and QVQ across multiple dimensions, including coarse-grained macro-expression classification, fine-grained micro-expression classification, and language generation. Our results highlight the effectiveness, interpretability, and adaptation potential of large models in micro-expression understanding. The code is available at: https://github.com/2308623956/MEGC2025.
Micro-expressions (MEs) are involuntary, low-intensity, and short-duration facial expressions that often reveal an individual's genuine thoughts and emotions. Most existing ME analysis methods rely on window-level classification with fixed window sizes and hard decisions, which limits their ability to capture the complex temporal dynamics of MEs. Although recent approaches have adopted video-level regression frameworks to address some of these challenges, interval decoding still depends on manually predefined, window-based methods, leaving the issue only partially mitigated. In this paper, we propose a prior-guided video-level regression method for ME analysis. We introduce a scalable interval selection strategy that comprehensively considers the temporal evolution, duration, and class distribution characteristics of MEs, enabling precise spotting of the onset, apex, and offset phases. In addition, we introduce a synergistic optimization framework, in which the spotting and recognition tasks share parameters except for the classification heads. This fully exploits complementary information, makes more efficient use of limited data, and enhances the model's capability. Extensive experiments on multiple benchmark datasets demonstrate the state-of-the-art performance of our method, with an STRS of 0.0562 on CAS(ME)3 and 0.2000 on SAMMLV. The code is available at https://github.com/zizheng-guo/BoostingVRME.
Major Depressive Disorder (MDD) is a prevalent and severe psychiatric disorder, and its detection remains challenging due to the complexity and variability of its symptoms. Traditional single-modality methods often fail to capture the full spectrum of depressive cues, which has led to the rise of multimodal methods. The ACM Multimedia 2025 ''Multimodal Personality-Aware Depression Detection Challenge'' (MPDD 2025) aims to advance the development of more accurate depression detection models by incorporating multimodal data. In this paper, we proposed a Multi-Level Segment Fusion Based on Adaptive Time-Window Selection (MSF-ATS) method for the MPDD-Elderly Track. To address the challenge of sparse and transient depressive symptoms, we fuse segment-level classifications to obtain subject-level classifications. An adaptive time-window selection based on mean class variance is employed to choose the window with the smallest variance for more stable detection results. Our method achieved an average score of 0.8576 on the MPDD 2025 official test set, significantly outperforming the baseline score of 0.6675.
Depression is an increasingly prevalent mental health issue worldwide, especially among the elderly, where effective early detection is crucial for timely intervention. While recent multimodal approaches demonstrate promise in leveraging visual and auditory cues for automatic depression recognition, existing methods often fail to extract fine-grained, depression-specific patterns from pre-extracted features and underutilize cross-modal interactions. To address these challenges, we propose DepFormer, a unified framework that incorporates a Bimodal Collaborative Transformer(BCT) for cross-modal representation learning and a personalized fusion module to enhance individual-specific modeling. The architecture comprises: (1) unimodal feature extraction, (2) bimodal collaborative representation learning via the BCT, (3) personalized feature fusion, and (4) final depression classification. Critically, the BCT employs symmetric bidirectional branches comprising an Audio-to-Video Transformer and a Video-to-Audio Transformer to enable mutual enhancement and complementary learning across modalities. Extensive experiments validate DepFormer's effectiveness, securing first place in the MPDD Challenge 2025 (Elderly Track), underscoring its strong practical potential.
We present a Collaborative Agent-Based Framework for Multi-Image Reasoning. Our approach tackles the challenge of interleaved multimodal reasoning across diverse datasets and task formats by employing a dual-agent system: a language-based PromptEngineer, which generates context-aware, task-specific prompts, and a VisionReasoner, a large vision-language model (LVLM) responsible for final inference. The framework is fully automated, modular, and training-free, enabling generalization across classification, question answering, and free-form generation tasks involving one or multiple input images. We evaluate our method on 18 diverse datasets from the 2025 MIRAGE Challenge (Track A), covering a broad spectrum of visual reasoning tasks including document QA, visual comparison, dialogue-based understanding, and scene-level inference. Our results demonstrate that LVLMs can effectively reason over multiple images when guided by informative prompts. Notably, Claude 3.7 achieves near-ceiling performance on challenging tasks such as TQA (99.13% accuracy), DocVQA (96.87%), and MMCoQA (75.28 ROUGE-L). We also explore how design choices-such as model selection, shot count, and input length-influence the reasoning performance of different LVLMs.
Due to their powerful image generation capabilities, diffusion-based adversarial example generation methods through image editing are rapidly gaining popularity. However, due to reliance on the discriminative capability of the diffusion model, these diffusion-based methods often struggle to generalize beyond conventional image classification tasks, such as in Deepfake detection. Moreover, traditional strategies for enhancing adversarial example transferability are challenging to adapt to these methods. To address these challenges, we propose a unified framework that seamlessly incorporates traditional transferability enhancement strategies into diffusion model-based adversarial example generation via image editing, enabling their application across a wider range of downstream tasks. Our method won first place in the ''1st Adversarial Attacks on Deepfake Detectors: A Challenge in the Era of AI-Generated Media'' competition at ACM MM25, which validates the effectiveness of our approach.
The rapid proliferation of AI-generated media, particularly hyper-realistic deepfakes, has underscored the critical need for robust detection systems to mitigate risks such as misinformation and identity theft. However, state-of-the-art deepfake detectors remain vulnerable to adversarial attacks-subtle perturbations designed to evade classification. To address this gap, we organized the Adversarial Attacks on Deepfake Detectors (AADD-2025) challenge, a competitive evaluation aimed at advancing methodologies to expose and strengthen weaknesses in deepfake detection models. The challenge tasked participants with generating adversarial examples capable of evading four diverse classifiers (including ResNet, DenseNet, and two blind models) while preserving structural similarity to original deepfakes. A dataset comprising 16 subsets of high- and low-quality deepfake images generated by GAN-based and diffusion models (e.g., StableDiffusion, StyleGAN3) was provided. Participants were evaluated using a weighted combination of Structural Similarity Index (SSIM) and attack success rates across all classifiers. Thirteen teams proposed innovative solutions leveraging techniques such as latent-space manipulation, ensemble gradient optimization, surrogate modeling, and frequency-domain perturbation. Top-performing approaches, including MR-CAS (1st place), Safe AI (2nd place), and RoMa (3rd place), achieved high SSIM scores (0.74-0.93) while successfully misleading classifiers. Notably, MR-CAS's latent diffusion model inversion strategy and Safe AI's consensus-orthogonal gradient weighting framework demonstrated superior transferability across architectures, including Vision Transformers. The challenge revealed critical insights: latent-space attacks outperformed pixel-level methods, ensemble-based strategies enhanced cross-model robustness, and adversarial perturbations optimized for both CNNs and transformers proved most effective. However, gaps persist in generalizing attacks across heterogeneous models and maintaining perceptual fidelity, highlighting the urgency of developing adaptive defenses and hybrid detection mechanisms. By fostering collaboration and innovation, AADD-2025 provides a benchmark for evaluating adversarial robustness in deepfake detection and underscores the need for resilient systems in the era of AI-generated media.
The rapid development of audio-driven talking head generators and advanced Text-To-Speech (TTS) models has led to more sophisticated temporal deepfakes. These advances highlight the need for robust methods capable of detecting and localizing deepfakes, even under novel, unseen attack scenarios. Current state-of-the-art deepfake detectors, while accurate, are often computationally expensive and struggle to generalize to novel manipulation techniques. To address these challenges, we propose multimodal approaches for the AV-Deepfake1M 2025 challenge. For the visual modality, we leverage handcrafted features to improve interpretability and adaptability. For the audio modality, we adapt a self-supervised learning (SSL) backbone coupled with graph attention networks to capture rich audio representations, improving detection robustness. Our approach strikes a balance between performance and real-world deployment, focusing on resilience and potential interpretability. On the AV-Deepfake1M++ dataset, our multimodal system achieves AUC of 92.78% for deepfake classification task and IoU of 0.3536 for temporal localization using only the audio modality.
The field of visual and audio generation is burgeoning with new state-of-the-art methods. This rapid proliferation of new techniques underscores the need for robust solutions for detecting synthetic content in videos. In particular, when fine-grained alterations via localized manipulations are performed in visual, audio, or both domains, these subtle modifications add challenges to the detection algorithms. This paper presents solutions for the problems of deepfake video classification and localization. The methods were submitted to the ACM 1M Deepfakes Detection Challenge, achieving the best performance in the temporal localization task and a top four ranking in the classification task for the TestA split of the evaluation dataset.
In this paper, we present PrivEdit, a zero-shot, interactive image privacy editing system specifically designed for automated sensitive information desensitization. As social networks and smart devices proliferate, the risk of unintended privacy leakage grows, driving demand for personalized, controllable protection tools. PrivEdit is powered by natural-language instructions and integrates a Recognize-Anything model for robust detection and classification of sensitive objects (e.g., faces, license plates, ID cards), followed by GroundingDINO and SAM for high-precision mask extraction. User intents are parsed and disambiguated via GPT-4o, enabling selective target confirmation and iterative refinement. Finally, our editing module performs localized edits-such as adjustable blurring, mosaicking, or replacement via generative editing. With support for multi-round feedback and real-time modification, PrivEdit seamlessly handles both pre-recorded images and live streams, making it ideal for social-media pre-publishing, privacy data desensitization in enterprise or healthcare contexts, and intelligent surveillance applications. By unifying detection, segmentation, intent parsing, and localized editing into one coherent interface, PrivEdit delivers an end-to-end solution for safeguarding visual data. Supplementary materials including the demo video and slides are available at: https://drive.google.com/file/d/13jFBmYgZgxYQLPIAqCeaQhcyzZTHpf7N/view?usp=sharing
Progress in understanding real-world canine vocal communication is constrained by datasets lacking scale and 'in-the-wild' diversity. We introduce DogSpeak, a large-scale public dataset of 77,202 Barkseqs (33.162 hours) from 156 dogs (5 breeds), uniquely sourced from online social media with accurate dog ID, sex, and breed labels. DogSpeak, one of the largest of its kind, addresses prior limitations. Benchmark tasks (sex, breed, individual dog recognition) demonstrate its utility and highlight how its inherent real-world challenges necessitate and foster research into more robust bioacoustic models, preprocessing, and feature representation.
Detecting hate speech in videos remains challenging due to the complexity of multimodal content and the lack of fine-grained annotations in existing datasets. We present HateClipSeg, a large-scale multimodal dataset with both video-level and segment-level annotations, comprising over 11,714 segments labeled as Normal or across five Offensive categories: Hateful, Insulting, Sexual, Violence, Self-Harm, along with explicit target victim labels. Our three-stage annotation process yields high inter-annotator agreement (Krippendorff's alpha = 0.817). We propose three tasks to benchmark performance: (1) Trimmed Hateful Video Classification, (2) Temporal Hateful Video Localization, and (3) Online Hateful Video Classification. Results highlight substantial gaps in current models, emphasizing the need for more sophisticated multimodal and temporally aware approaches. The HateClipSeg dataset are publicly available at https://github.com/Social-AI-Studio/HateClipSeg.git.
Osteoporosis has emerged as a significant global public health challenge, affecting more than 200 million people worldwide. Precise screening through bone mineral density (BMD) assessment is essential for timely intervention. Although dual-energy X-ray absorptiometry (DXA) serves as the gold standard for BMD evaluation, its widespread application is hampered by high costs and limited accessibility. In recent years, deep learning has demonstrated great potential in osteoporosis classification and predicting BMD from X-ray and CT scans. However, existing public datasets have notable limitations, particularly the absence of paired multimodal imaging datasets for cross-modal modeling and the lack of accurate BMD annotations required for osteoporosis-specific research and clinical model optimization. To address these issues, we introduce the Lumbar Multimodal Osteoporosis Screening dataset (LUMOS), the first multimodal dataset specifically designed for lumbar osteoporosis screening. LUMOS integrates clinical data from 803 patients, including 1,620 anteroposterior/lateral lumbar X-rays with BMD values and T-scores, comprehensive demographic information, and 280 lumbar CT scans. The advent of LUMOS is expected to propel forward research on automated osteoporosis classification, BMD prediction, and other related tasks. Its standardized and multimodal nature fills critical gaps in lumbar osteoporosis data, providing high-quality data support for the development and validation of medical AI algorithms in the early detection of osteoporosis. The dataset is available at https://keyueshi.github.io/LUMOS/.
Food image classification models are crucial for dietary management applications because they reduce the burden of manual meal logging. However, most publicly available datasets for training such models rely on web-crawled images, which often differ from users' real-world meal photos. In this work, we present FoodLogAthl-218, a food image dataset constructed from real-world meal records collected through the dietary management application FoodLog Athl. The dataset contains 6,925 images across 218 food categories, with a total of 14,349 bounding boxes. Rich metadata, including meal date and time, anonymized user IDs, and meal-level context, accompany each image. Unlike conventional datasets-where a predefined class set guides web-based image collection-our data begins with user-submitted photos, and labels are applied afterward. This yields greater intra-class diversity, a natural frequency distribution of meal types, and casual, unfiltered images intended for personal use rather than public sharing. In addition to (1) a standard classification benchmark, we introduce two FoodLog-specific tasks: (2) an incremental fine-tuning protocol that follows the temporal stream of users' logs, and (3) a context-aware classification task where each image contains multiple dishes, and the model must classify each dish by leveraging the overall meal context. We evaluate these tasks using large multimodal models (LMMs). The dataset is publicly available at https://huggingface.co/datasets/FoodLog/FoodLogAthl-218.
AudioSet is a widely used benchmark in the audio research community and has significantly advanced various audio-related tasks. However, persistent issues with label accuracy and completeness remain critical bottlenecks that limit performance in downstream applications. To address the aforementioned challenges, we propose a three-stage reannotation framework that harnesses general-purpose audio-language foundation models to systematically improve the label quality of AudioSet. The framework employs a cross-modal prompting strategy, inspired by the concept of prompt chaining, wherein prompts are sequentially composed to execute subtasks (audio comprehension, label synthesis, and semantic alignment). Leveraging this framework, we construct a high-quality, structured relabeled version of AudioSet-R. Extensive experiments conducted on representative audio classification models-including AST, PANNs, SSAST, and AudioMAE-consistently demonstrate substantial performance improvements, thereby validating the generalizability and effectiveness of the proposed approach in enhancing label reliability.The code is publicly available at: https://github.com/colaudiolab/AudioSet-R.