We introduce MM-Mixing, a multi-modal mixing alignment framework for 3D understanding. MM-Mixing applies mixing-based methods to multi-modal data, preserving and optimizing cross-modal connections while enhancing diversity and improving alignment across modalities. Our proposed two-stage training pipeline combines feature-level and input-level mixing to optimize the 3D encoder. The first stage employs feature-level mixing with contrastive learning to align 3D features with their corresponding modalities. The second stage incorporates both feature-level and input-level mixing, introducing mixed point cloud inputs to further refine 3D feature representations. MM-Mixing enhances intermodality relationships, promotes generalization, and ensures feature consistency while providing diverse and realistic training samples. We demonstrate that MM-Mixing significantly improves baseline performance across various learning scenarios, including zero-shot 3D classification, linear probing 3D classification, and cross-modal 3D shape retrieval. Notably, we improved the zero-shot classification accuracy on ScanObjectNN from 51.3% to 61.9%, and on Objaverse-LVIS from 46.8% to 51.4%. Our findings highlight the potential of multi-modal mixing-based alignment to significantly advance 3D object recognition and understanding while remaining straightforward to implement and integrate into existing frameworks.
论文检索
输入标题、作者或关键词，从 1,144 篇学术成果中精准定位
This paper develops a Versatile and Honest vision language Model (VHM) for remote sensing image analysis. VHM is built on a large-scale remote sensing image-text dataset with rich-content captions (VersaD), and an honest instruction dataset comprising both factual and deceptive questions (HnstD). Unlike prevailing remote sensing image-text datasets, in which image captions focus on a few prominent objects and their relationships, VersaD captions provide detailed information about image properties, object attributes, and the overall scene. This comprehensive captioning enables VHM to thoroughly understand remote sensing images and perform diverse remote sensing tasks. Moreover, different from existing remote sensing instruction datasets that only include factual questions, HnstD contains additional deceptive questions stemming from the non-existence of objects. This feature prevents VHM from producing affirmative answers to nonsense queries, thereby ensuring its honesty. In our experiments, VHM significantly outperforms various vision language models on common tasks of scene classification, visual question answering, and visual grounding. Additionally, VHM achieves competent performance on several unexplored tasks, such as building vectorizing, multi-label classification and honest question answering.
Procedure Knowledge Decoupled Distillation Strategy for Procedure Planning in Instructional Videos
PDF ↗Procedure planning in instructional videos, producing a structured and plannable action sequence facilitating the transition from the start to the goal states, has achieved significant progress. The dominant single-branch non-autoregressive planning paradigm guides action sequence generation through action labels, overlooking the limitation of the absence of intermediate visual information. Hence, we introduce the procedure knowledge decoupled distillation strategy to address the above issue. This innovative strategy deliberately lets the teacher model see the real visual information among the start and goal states to enhance its action semantic understanding and relationship modeling ability, producing the potential probability distribution containing the real action class and other action classes that may occur. Accordingly, we introduce a decoupled intermediate information knowledge distillation loss, which comprises single action knowledge distillation and sequence distribution knowledge distillation for the student model. The former improves the student model's precise inference ability for individual actions by transferring knowledge of a single action target category using binary classification loss. Conversely, the latter uses MSE loss to constrain the student model to learn the action sequence probability distribution from the teacher model, thereby enhancing the student model's global planning capability. Extensive experiments on three datasets demonstrate that our strategy can improve the performance of multiple weakly supervised models, achieving promising procedure knowledge modeling ability and plug-and-play flexibility.
DigitalLLaVA: Incorporating Digital Cognition Capability for Physical World Comprehension in Multimodal LLMs
PDF ↗Multimodal Large Language Models (MLLMs) have shown remarkable cognitive capabilities in various cross-modal tasks.However, existing MLLMs struggle with tasks that require physical digital cognition, such as accurately reading an electric meter or pressure gauge. This limitation significantly reduces their effectiveness in practical applications like industrial monitoring and home energy management, where digital sensors are not feasible. For humans, physical digits are artificially defined quantities presented on specific carriers, which require training to recognize. As existing MLLMs are only pre-trained in the manner of object recognition, they fail to comprehend the relationship between digital carriers and their reading. To this end, referring to human behavior, we propose a novel DigitalLLaVA method to explicitly inject digital cognitive abilities into MLLMs in a two-step manner. In the first step, to improve the MLLM's understanding of physical digit carriers, we propose a digit carrier mapping method. This step utilizes object-level text-image pairs to enhance the model's comprehension of objects containing physical digits. For the second step, unlike previous methods that rely on sequential digital prediction or digit regression, we propose a 32 bit floating point simulation approach that treats digit prediction as a whole. Using digit-level text-image pairs, we train three float heads to predict 32-bit floating-point numbers using 0/1 binary classification. This step significantly reduces the search space, making the prediction process more robust and straightforward. Being simple but effective, our method can identify very precise metrics (i.e., accurate to ±0.001) and provide floating-point results, showing its applicability in digital carrier domains.
Similar Modality Enhancement and Action Consistency Learning for Weakly Supervised Temporal Action Localization
PDF ↗Weakly-supervised temporal action localization (WTAL) aims to identify and localize action instances in untrimmed videos using only video-level labels. Existing methods typically rely on original features from frozen pre-trained encoders designed for trimmed action classification (TAC) tasks, which inevitably introduces task discrepancy. Additionally, these methods often overlook the importance of considering action consistency from multiple perspectives, specifically the consistency in action processes and action semantics, both of which are crucial for the model's understanding of actions. To address these issues, we propose a novel WTAL method based on similar modality enhancement and action consistency learning (SEAL). First, we construct global descriptors for each action category, and use the pseudo-labels generated based on these descriptors to guide the model in learning more consistent representations, thereby mitigating task discrepancy. Second, we design two types of losses to achieve action consistency learning: process consistency loss, which penalizes candidate proposals that deviate from the action center to ensure the completeness of the action process, and semantic consistency loss, which employs local descriptors to help proposals of the same action category (especially those with apparent semantic confusion) learn similar feature distributions. Extensive experiments on the THUMOS14 and ActivityNet datasets demonstrate the superior performance of the proposed method compared to state-of-the-art methods.
Deep learning has excelled in medical image classification, but its clinical application is limited by poor interpretability. Capsule networks, known for encoding hierarchical relationships and spatial features, show potential in addressing this issue. Nevertheless, traditional capsule networks often underperform due to their shallow structures, and deeper variants lack hierarchical architectures, thereby compromising interpretability. This paper introduces a novel capsule network, ParseCaps, which utilizes the sparse axial attention routing and parse convolutional capsule layer to form a parse-tree-like structure, enhancing both depth and interpretability. Firstly, sparse axial attention routing optimizes connections between child and parent capsules, as well as emphasizes the weight distribution across instantiation parameters of parent capsules. Secondly, the parse convolutional capsule layer generates capsule predictions aligning with the parse tree. Finally, based on the loss design that is effective whether concept ground truth exists or not, ParseCaps advances interpretability by associating each dimension of the global capsule with a comprehensible concept, thereby facilitating clinician trust and understanding of the model's classification results. Experimental results on three medical datasets show that ParseCaps not only outperforms other capsule network variants in classification accuracy and robustness, but also provides interpretable explanations, regardless of the availability of concept labels.
With the rapid advancement of 3D scanning technology, point clouds have become a crucial data type in computer vision and machine learning. However, learning robust representations for point clouds remains a significant challenge due to their irregularity and sparsity. In this paper, we propose a novel Dual Manifold Regularization (DMR) framework that makes full use of the properties of positive and negative curvature in manifolds to improve the representation of point clouds. Specifically, we leverage DMR based on hyperbolic and hyperspherical manifolds to address the limitations of traditional single-manifold regularization techniques, including inadequate generalization ability and adaptability to data diversity, as well as the difficulty of capturing complex relationships between data. To begin, we utilize the tree-like structure of the hyperbolic manifold to model the part-whole hierarchical relationships within point clouds. This allows for a more comprehensive representation of the data, improving the model's capability to understand complex shapes. Additionally, we construct positive samples through topological consistency augmentation and employ contrastive learning techniques in the hyperspherical manifold to capture more discriminative features within the data. Our experimental results show that our method outperforms traditional supervised learning and single-manifold regularization techniques in point cloud analysis. Specifically, for shape classification, DMR achieves a new State-Of-The-Art (SOTA) performance with 94.8% Overall Accuracy (OA) on ModelNet40 and 90.7% OA on ScanObjectNN, surpassing the recent SOTA model without increasing the baseline parameters.
Emotion recognition based on body movements is vital in human-computer interaction. However, existing emotion recognition methods predominantly focus on enhancing classification accuracy, often neglecting the provision of textual explanations to justify their classifications. In this paper, we propose an Emotion-Action Interpreter powered by LargeLanguage Model (EAI-LLM), which not only recognizes emotions but also generates textual explanations by treating 3D body movement data as unique input tokens within large language models (LLMs). Specifically, we propose a multi-granularity skeleton tokenizer designed for LLMs, which separately extracts spatio-temporal tokens and semantic tokens from the skeleton data. This approach allows LLMs to generate more nuanced classification descriptions while maintaining robust classification performance. Furthermore, we treat the skeleton sequence as a specific language and propose a unified skeleton token module. This module leverages the extensive background knowledge and language processing capabilities of LLMs to address the challenges of joint training on heterogeneous datasets, thereby significantly enhancing recognition accuracy on individual datasets. Experimental results demonstrate that our model achieves recognition accuracy comparable to existing methods. More importantly, with the support of background knowledge from LLMs, our model can generate detailed emotion descriptions based on classification results, even when trained on a limited amount of labeled skeleton data.
Existing Weakly Supervised Semantic Segmentation (WSSS) relies on the CNN-based Class Activation Map (CAM) and Transformer-based self-attention map to generate class-specific masks for semantic segmentation. However, CAM and self-attention maps usually cause incomplete segmentation due to classification bias issue. To address this issue, we propose a Multi-Label Prototype Visual Spatial Search (MuP-VSS) method with a spatial query mechanism, which learns a set of learnable class token vectors as queries to search the similarity visual tokens from image patch tokens. Specifically, MuP-VSS consists of two key components: multi-label prototype representation and multi-label prototype optimization. The former designs a global embedding to learn the global tokens from the images, and then proposes a Prototype Embedding Module (PEM) to interact with patch tokens to understand the local semantic information. The latter utilizes the exclusivity and consistency principles of the multi-label prototypes to design three prototype losses to optimize them, which contain cross-class prototype (CCP) contrastive loss, cross-image prototype (CIP) contrastive loss, and patch-to-prototype (P2P) consistency loss. CCP loss models exclusivity of multi-label prototypes learned from a single image to enhance the discriminative properties of each class better. CCP loss learns the consistency of the same class-specific prototypes extracted from multiple images to enhance the semantic consistency. P2P loss is proposed to control the semantic response of the prototype to the image patches. Experimental results on Pascal VOC 2012 and MS COCO show that MuP-VSS significantly outperforms recent methods and achieves state-of-the-art performance.
As a globally celebrated sport, soccer has attracted widespread interest from fans over the world. This paper aims to develop a comprehensive multi-modal framework for soccer video understanding.Specifically, we make the following contributions in this paper:(i) we introduce **SoccerReplay-1988**, the largest multi-modal soccer dataset to date, featuring videos and detailed annotations from 1,988 complete matches, with an automated annotation pipeline;(ii) we present the first visual-language foundation model in the soccer domain, **MatchVision**, which leverages spatiotemporal information across soccer videos and excels in various downstream tasks;(iii) we conduct extensive experiments and ablation studies on action classification, commentary generation, and multi-view foul recognition,and demonstrate state-of-the-art performance on all of them, substantially outperforming existing models, which has demonstrated the superiority of our proposed data and model. We believe that this work will offer a standard paradigm for sports understanding research. The code and model will be publicly available for reproduction.
Bringing CLIP to the Clinic: Dynamic Soft Labels and Negation-Aware Learning for Medical Analysis
PDF ↗The development of large-scale image-text pair datasets has significantly advanced self-supervised learning in Vision-Language Processing (VLP). However, directly applying general-domain architectures such as CLIP to medical data presents challenges, particularly in handling negations and addressing the inherent data imbalance of medical datasets. To address these issues, we propose a novel approach that integrates clinically-enhanced dynamic soft labels and medical graphical alignment, thereby improving clinical comprehension and improving the applicability of contrastive loss in medical contexts. Furthermore, we introduce negation-based hard negatives to deepen the model's understanding of the complexities of clinical language. Our approach is easily integrated into medical CLIP training pipeline and achieves state-of-the-art performance across multiple tasks, including zero-shot, fine-tuned classification and report retrieval. To comprehensively evaluate our model's capacity in understanding clinical language, we introduce CXR-Align, a benchmark uniquely designed to evaluate the understanding of negation and clinical information within chest X-ray (CXR) datasets. Experimental results demonstrate that our proposed methods are straightforward to implement and generalize effectively across contrastive learning frameworks, enhancing medical VLP capabilities and advancing clinical language understanding in medical imaging.
Recent multi-teacher distillation methods have unified the encoders of multiple foundation models into a single encoder, achieving competitive performance on core vision tasks like classification, segmentation, and depth estimation. This led us to ask: Could similar success be achieved when the pool of teachers also includes vision models specialized in diverse tasks across both 2D and 3D perception? In this paper, we define and investigate the problem of heterogeneous teacher distillation, or co-distillation, a challenging multi-teacher distillation scenario where teacher models vary significantly in both (a) their design objectives and (b) the data they were trained on. We explore data-sharing strategies and teacher-specific encoding, and introduce DUNE, a single encoder excelling in 2D vision, 3D understanding, and 3D human perception. Our model achieves performance comparable to that of its larger teachers, sometimes even outperforming them, on their respective tasks. Notably, DUNE surpasses MASt3R in Map-free Visual Relocalization with a much smaller encoder.
How can we troubleshoot a deep visual model, i.e., understand why it makes certain mistakes and further take action to correct its behavior? We design a Model Diagnosis and Correction system (MDC), an automated framework that analyzes the pattern of errors, proposes candidate causes of attributes, conducts hypothesis testing via attribute editing, and ultimately generates counterfactual training samples to improve the performance of the model. Unlike previous methods, in addition to the linguistic attributes, our method also incorporates the analysis for implicit causal attributes, those cannot to be accurately described by language. To achieve this, we propose an image editing module capable of leveraging both implicit and linguistic attributes to generate counterfactual images depicting error patterns and further experimentally validate causality relationships. Lastly, we enrich the training set with synthetic samples depicting verified causal attributes and retrain the model, further boosting accuracy and robustness. Extensive experiments on fine-grained classification and face security applications demonstrate the superiority of our approach in model diagnosis and correction. Specifically, we achieve an average relative improvement of 62.01% in HTER for face security application over state-of-the-art methods.
Multi-label classification is crucial for comprehensive image understanding, yet acquiring accurate annotations is challenging and costly. To address this, a recent study suggests exploiting unsupervised multi-label classification leveraging CLIP, a powerful vision-language model. Despite CLIP's proficiency, it suffers from view-dependent predictions and inherent bias, limiting its effectiveness. We propose a novel method that addresses these issues by leveraging multiple views near target objects, guided by Class Activation Mapping (CAM) of the classifier, and debiasing pseudo-labels derived from CLIP predictions. Our Classifier-guided CLIP Distillation (CCD) enables selecting multiple local views without extra labels and debiasing predictions to enhance classification performance. Experimental results validate our method's superiority over existing techniques across diverse datasets. The code is available at https://github.com/k0u-id/CCD.
Concept-based models can map black-box representations to human-understandable concepts, which makes the decision-making process more transparent and then allows users to understand the reason behind predictions. However, domain-specific concepts often impact the final predictions, which subsequently undermine the model generalization capabilities, and prevent the model from being used in high-stake applications. In this paper, we propose a novel Language-guided Concept-Erasing (LanCE) framework. In particular, we empirically demonstrate that pre-trained vision-language models (VLMs) can approximate distinct visual domain shifts via domain descriptors while prompting large Language Models (LLMs) can easily simulate a wide range of descriptors of unseen visual domains. Then, we introduce a novel plug-in domain descriptor orthogonality (DDO) regularizer to mitigate the impact of these domain-specific concepts on the final predictions. Notably, the DDO regularizer is agnostic to the design of conceptbased models and we integrate it into several prevailing models. Through evaluation of domain generalization on four standard benchmarks and three newly introduced benchmarks, we demonstrate that DDO can significantly improve the out-of-distribution (OOD) generalization over the previous state-of-the-art concept-based models. Our code is available at https://github.com/joeyz0z/LanCE.
A fundamental aspect of compositional reasoning in a video is associating people and their actions across time. Recent years have seen great progress in general-purpose vision/video models and a move towards long-video understanding. While exciting, we take a step back and ask: are today's models good at compositional reasoning on short videos? To this end, we introduce VELOCITI, a benchmark to study Video-LLMs by disentangling and assessing the comprehension of agents, actions, and their associations across multiple events. We adopt the Video-Language Entailment setup and propose StrictVLE that requires correct classification (rather than ranking) of the positive and negative caption. We evaluate several models and observe that even the best, LLaVA-OneVision (44.5%) and Gemini-1.5-Pro (49.3%), are far from human accuracy at 93.0%. Results show that action understanding lags behind agents, and negative captions created using entities appearing in the video perform worse than those obtained from pure text manipulation. We also present challenges with ClassicVLE and multiple-choice (MC) evaluation, strengthening our preference for StrictVLE. Finally, we validate that our benchmark requires visual inputs of multiple frames making it ideal to study video-language compositional reasoning.
Vision-Language Models (VLMs) trained with contrastive loss have achieved significant advancements in various vision and language tasks. However, the global nature of the contrastive loss makes VLMs focus predominantly on foreground objects, neglecting other crucial information in the image, which limits their effectiveness in downstream tasks. To address these challenges, we propose COSMOS: CrOSs-MOdality Self-distillation for vision-language pre-training that integrates a novel text-cropping strategy and cross-attention module into a self-supervised learning framework. We create global and local views of images and texts (i.e., multi-modal augmentations), which are essential for self-distillation in VLMs. We further introduce a cross-attention module, enabling COSMOS to learn comprehensive cross-modal representations optimized via a cross-modality self-distillation loss. COSMOS consistently outperforms previous strong baselines on various zero-shot downstream tasks, including retrieval, classification, and semantic segmentation. Additionally, it surpasses CLIP-based models trained on larger datasets in visual perception and contextual understanding tasks. Code is available at https://github.com/ExplainableML/cosmos.
We propose a new "Unbiased through Textual Description (UTD)" video benchmark based on unbiased subsets of existing video classification and retrieval datasets to enable a more robust assessment of video understanding capabilities. Namely, we tackle the problem that current video benchmarks may suffer from different representation biases, e.g., object bias or single-frame bias, where mere recognition of objects or utilization of only a single frame is sufficient for correct prediction. We leverage VLMs and LLMs to analyze and debias benchmarks from such representation biases. Specifically, we generate frame-wise textual descriptions of videos, filter them for specific information (e.g. only objects) and leverage them to examine representation biases across three dimensions: 1) concept bias -- determining if a specific concept (e.g., objects) alone suffice for prediction; 2) temporal bias -- assessing if temporal information contributes to prediction; and 3) common sense vs. dataset bias -- evaluating whether zero-shot reasoning or dataset correlations contribute to prediction. We conduct a systematic analysis of 12 popular video classification and retrieval datasets and create new object-debiased test splits for these datasets. Moreover, we benchmark 30 state-of-the-art video models on original and debiased splits and analyze biases in the models. To facilitate the future development of more robust video understanding benchmarks and models, we release: "UTD-descriptions", a dataset with our rich structured descriptions for each dataset, and "UTD-splits", a dataset of object-debiased test splits.
Motion expression video segmentation aims to segment objects based on input motion descriptions. Compared with traditional referring video object segmentation, it focuses on motion and multi-object expressions and is more challenging. Previous works achieved it by simply injecting text information into the video instance segmentation (VIS) model. However, this requires retraining the entire model and optimization is difficult. In this work, we propose DMVS, a simple framework constructed on the existing query-based VIS model, emphasizing decoupling the task into video instance segmentation and motion expression understanding. Firstly, we use a frozen video instance segmenter to extract object-specific contexts and convert them into frame-level and video-level queries. Secondly, we interact two levels of queries with static and motion cues, respectively, to further encode visually enhanced motion expressions. Furthermore, we propose a novel query initialization strategy that uses video queries guided by classification priors to initialize motion queries, greatly reducing the difficulty of optimization. Without bells and whistles, DMVS achieves state-of-the-art performance on the MeViS dataset at a lower training cost. Extensive experiments verify the effectiveness and efficiency of our framework.
AdaCM^2: On Understanding Extremely Long-Term Video with Adaptive Cross-Modality Memory Reduction
PDF ↗The advancements in large language models (LLMs) have propelled the improvement of video understanding tasks by incorporating LLMs with visual models. However, most existing LLM-based models (e.g., VideoLLaMA, VideoChat) are constrained to processing short-duration videos. Recent attempts to understand long-term videos by extracting and compressing visual features into a fixed memory size. Nevertheless, those methods leverage only visual modality to merge video tokens and overlook the correlation between visual and textual queries, leading to difficulties in effectively handling complex question-answering tasks. To address the challenges of long videos and complex prompts, we propose AdaCM^2, which, for the first time, introduces an adaptive cross-modality memory reduction approach to video-text alignment in an auto-regressive manner on video streams. Our extensive experiments on various video understanding tasks, such as video captioning, video question answering, and video classification, demonstrate that AdaCM^2 achieves state-of-the-art performance across multiple datasets while significantly reducing memory usage. Notably, it achieves a 4.5% improvement across multiple tasks in the LVU dataset with a GPU memory consumption reduction of up to 65%.