Fetal ultrasound screening is a uniquely complex diagnostic task involving the simultaneous assessment of multiple fetal organs—each with its own anatomical and clinical context—within a single examination. Automating report generation for such cases poses a significant challenge: unlike existing methods that focus on single-organ radiology tasks (e.g., chest X-rays), fetal ultrasound requires reasoning over a structured, multiple-to-multiple setting, i.e., multi-organ images corresponding to a multi-section report. In this paper, we introduce FetusR, the first large-scale dataset for multi-organ fetal ultrasound reporting, containing 15,594 real-world cases with rich organ-wise annotations. To address the intrinsic image-report alignment, we propose Organ-Aware Routing Mixture-of-Retrieval Augmented Generation (ORM-RAG) inspired by the Mixture-of-Experts paradigm. Our method decomposes the complex alignment problem into multiple one-to-one sub-retrieval tasks. Specifically, ORM-RAG integrates (1) an organ-aware mixture-of-retrieval module that partitions the retrieval space into organ-specific corpora for independent retrieval, and (2) a dynamic routing mechanism that selectively aggregates high-confidence organ-specific reports while filtering uncertain ones. Extensive experiments demonstrate that our method significantly outperforms state-of-the-art baselines across both textual similarity and clinical accuracy metrics. Our work opens a new direction for long-form, structured report generation in real-world, multi-organ medical imaging scenarios.

High-resolution image inputs allow Large Vision-Language Models (LVLMs) to capture finer visual details, improving comprehension. However, the increased training and computational costs associated with such inputs pose significant challenges. A common approach to mitigate these costs involves slicing the input into uniform patches using sliding windows, each aligned with the vision encoder’s input size. While efficient, this method fragments the input, disrupting the continuity of context, which negatively impacts cross-patch perception tasks. To address these limitations, we propose HiRes-LLaVA, a novel framework designed to efficiently process high-resolution inputs of any size without altering the original contextual and geometric information. HiRes-LLaVA introduces two key components: (i) a SliceRestore Adapter (SRA) that reconstructs sliced patches into their original form, enabling efficient extraction of both global and local features through down-up-sampling and convolutional layers, and (ii) a Self-Mining Sampler (SMS) that compresses visual tokens based on internal relationships, preserving original context and positional information while reducing training overhead. To assess the ability of handling context fragmentation, we construct a new benchmark, EntityGrid-QA, consisting of edge-related tasks. Extensive experiments demonstrate the superiority of HiRes-LLaVA on both existing public benchmarks and EntityGrid-QA. For example, with SRA, our method achieves a performance improvement of ∼ 12% over state-of-the-art LVLMs in addressing fragmentation issues. Additionally, our SMS outperforms other visual token downsamplers, while offering high data efficiency.

Multimodal large language models (MLLMs) are broadly empowering various fields. Despite their advancements, the explainability of MLLMs remains less explored, hindering deeper understanding, model credibility, and effective visualization. Unlike conventional vision models (e.g., CNNs, ViTs, CLIP) that produce a single output, MLLMs generate sequences of tokens progressively, where each generated token depends on the previous context. Therefore, earlier context tokens can introduce redundant activations that interfere with the explanation of later tokens beyond their original information. Existing studies often overlook this issue, but our observations reveal that these redundant correlations can significantly hurt the reliability of explanations. To address this, we propose an estimated causal inference method to mitigate the interference of context to achieve high-quality MLLM explanation, with a novel rank Gaussian filter to further reduce activation noises. We term this method Token Activation Map (TAM) to highlight the consideration of interactions between tokens. TAM also indicates that it excels at explaining multiple tokens of MLLM, which is different from the Class Activation Map (CAM) for a single prediction. Our TAM method significantly outperforms existing SoTA methods, showcasing high-quality visualization results that can be utilized for various scenarios, such as object localization, failure case analysis, video visualization, MLLMs visual comparison, and model understanding (e.g., color, shape, action, location, visual reasoning, multi-turn conversation, etc). The code is available at github.com/xmed-lab/TAM.

Abstract Recent efforts to use natural language for interpretable driving focus mainly on planning, neglecting perception tasks. In this paper, we address this gap by introducing ROLISP (Risk Object Localization and Intention and Suggestion Prediction), which towards interpretable risk object detection and suggestion for ego car motions. Accurate ROLISP implementation requires extensive reasoning to identify critical traffic objects and infer their intentions, prompting us to explore the capabilities of multimodal large language models (MLLMs). However, the limited perception performance of CLIP-ViT vision encoders in existing MLLMs struggles with capturing essential visual perception information, e.g., high-resolution, multi-scale and visual-related inductive biases, which are important for autonomous driving. Addressing these challenges, we introduce HiLM-D, a resource-efficient framework that enhances visual information processing in MLLMs for ROLISP. Our method is motivated by the fact that the primary variations in autonomous driving scenarios are the motion trajectories rather than the semantic or appearance information (e.g., the shapes and colors) of objects. Hence, the visual process of HiLM-D is a two-stream framework: (i) a temporal reasoning stream, receiving low-resolution dynamic video content, to capture temporal semantics, and (ii) a spatial perception stream, receiving a single high-resolution frame, to capture holistic visual perception-related information. The spatial perception stream can be made very lightweight by a well-designed P-Adapter, which is lightweight, training-efficient, and easily integrated into existing MLLMs. Experiments on the DRAMA-ROLISP dataset show HiLM-D’s significant improvements over current MLLMs, with a $$3.7\%$$ 3.7 % in BLEU-4 for captioning and $$8.7\%$$ 8.7 % in mIoU for detection. Further tests on the Shikra-RD dataset confirm our method’s generalization capabilities. The DRAMA-ROLISP is available at https://github.com/xmed-lab/HiLM-D .

Early weakly supervised video grounding (WSVG) methods often struggle with incomplete boundary detection due to the absence of temporal boundary annotations. To bridge the gap between video-level and boundary-level annotations, explicit supervision methods (i.e., generating pseudo-temporal boundaries for training) have achieved great success. However, data augmentation in these methods might disrupt critical temporal information, yielding poor pseudo-temporal boundaries. In this paper, we propose a new perspective that maintains the integrity of the original temporal content while introducing more valuable information for expanding the incomplete boundaries. To this end, we proposeETC(ExpandthenClarify), first using the additional information to expand the initial incomplete pseudo-temporal boundaries, and subsequently refining these expanded ones to achieve precise boundaries. Motivated by video continuity, i.e., visual similarity across adjacent frames, we use powerful multi-modal large language models (MLLMs) to annotate each frame within the initial pseudo-temporal boundaries, yielding more comprehensive descriptions for expanded boundaries. To further clarify the noise in expanded boundaries, we combine mutual learning with a tailored proposal-level contrastive objective to use a learnable approach to harmonize a balance between incomplete yet clean (initial) and comprehensive yet noisy (expanded) boundaries for more precise ones. Experiments demonstrate the superiority of our method on two challenging WSVG datasets.

The rise of multimodal large language models (MLLMs) has spurred interest in language-based driving tasks. However, existing research typically focuses on limited tasks and often omits key multi-view and temporal information which is crucial for robust autonomous driving. To bridge these gaps, we introduce NuInstruct, a novel dataset with 91K multi-view video-QA pairs across 17 subtasks, where each task demands holistic information (e.g., temporal, multi-view, and spatial), significantly elevating the challenge level. To obtain NuInstruct, we propose a novel SQL-based method to generate instruction-response pairs automatically, which is inspired by the driving logical progression of humans. We further present BEV-InMLLM, an end-to-end method for efficiently deriving instruction-aware Bird's-Eye-View (BEV) features, language-aligned for large language models. BEV-InMLLM integrates multi-view, spatial awareness, and temporal semantics to enhance MLLMs' capabilities on NuInstruct tasks. Moreover, our proposed BEV injection module is a plug-and-play method for existing MLLMs. Our experiments on NuInstruct demonstrate that BEV-InMLLM significantly outperforms existing MLLMs, e.g. 9% improvement on various tasks. We release our NuInstruct at https://github.com/xmed-lab/NuInstruct.

Cone beam computed tomography (CBCT) is an important imaging technology widely used in medical scenarios, such as diagnosis and preoperative planning. Using fewer projection views to reconstruct CT, also known as sparse-view reconstruction, can reduce ionizing radiation and further benefit interventional radiology. Compared with sparse-view reconstruction for traditional parallel/fan-beam CT, CBCT reconstruction is more challenging due to the increased dimensionality caused by the measurement process based on cone-shaped X-ray beams. As a 2D-to-3D reconstruction problem, although implicit neural representations have been introduced to enable efficient training, only local features are considered and different views are processed equally in previous works, resulting in spatial inconsistency and poor performance on complicated anatomies. To this end, we propose C2RV by leveraging explicit multi-scale volumetric representations to enable cross-regional learning in the 3D space. Additionally, the scale-view cross-attention module is introduced to adaptively aggregate multi-scale and multi-view features. Extensive experiments demonstrate that our C2RV achieves consistent and significant improvement over previous state-of-the-art methods on datasets with diverse anatomy. Code is available at https://github.com/xmed-lab/C2RV-CBCT.

Weakly supervised temporal action localization (WTAL) aims to classify and localize temporal boundaries of actions for the video, given only video-level category labels in the training datasets. Due to the lack of boundary information during training, existing approaches formulate WTAL as a classification problem, i.e., generating the temporal class activation map (T-CAM) for localization. However, with only classification loss, the model would be suboptimized, i.e., the action-related scenes are enough to distinguish different class labels. Regarding other actions in the action-related scene (i.e., the scene same as positive actions) as co-scene actions, this suboptimized model would misclassify the co-scene actions as positive actions. To address this misclassification, we propose a simple yet efficient method, named bidirectional semantic consistency constraint (Bi-SCC), to discriminate the positive actions from co-scene actions. The proposed Bi-SCC first adopts a temporal context augmentation to generate an augmented video that breaks the correlation between positive actions and their co-scene actions in the inter-video. Then, a semantic consistency constraint (SCC) is used to enforce the predictions of the original video and augmented video to be consistent, hence suppressing the co-scene actions. However, we find that this augmented video would destroy the original temporal context. Simply applying the consistency constraint would affect the completeness of localized positive actions. Hence, we boost the SCC in a bidirectional way to suppress co-scene actions while ensuring the integrity of positive actions, by cross-supervising the original and augmented videos. Finally, our proposed Bi-SCC can be applied to current WTAL approaches and improve their performance. Experimental results show that our approach outperforms the state-of-the-art methods on THUMOS14 and ActivityNet. The code is available at https://github.com/lgzlIlIlI/BiSCC.

Due to the lack of temporal annotation, current Weakly-supervised Temporal Action Localization (WTAL) methods are generally stuck into over-complete or incomplete localization. In this paper, we aim to leverage the text information to boost WTAL from two aspects, i.e., (a) the discriminative objective to enlarge the inter-class difference, thus reducing the over-complete; (b) the generative objective to enhance the intra-class integrity, thus finding more complete temporal boundaries. For the discriminative objective, we propose a Text-Segment Mining (TSM) mechanism, which constructs a text description based on the action class label, and regards the text as the query to mine all class-related segments. Without the temporal annotation of actions, TSM compares the text query with the entire videos across the dataset to mine the best matching segments while ignoring irrelevant ones. Due to the shared sub-actions in different categories of videos, merely applying TSM is too strict to neglect the semantic-related segments, which results in incomplete localization. We further introduce a generative objective named Video-text Language Completion (VLC), which focuses on all semantic-related segments from videos to complete the text sentence. We achieve the state-of-the-art performance on THUMOS14 and ActivityNetl.3. Surprisingly, we also find our proposed method can be seamlessly applied to existing methods, and improve their performances with a clear margin. The code is available at https://github.com/lgzlIlIlI/Boosting-WTAL.

Echocardiogram video segmentation plays an important role in cardiac disease diagnosis. This paper studies the unsupervised domain adaption (UDA) for echocardiogram video segmentation, where the goal is to generalize the model trained on the source domain to other unlabelled target domains. Existing UDA segmentation methods are not suitable for this task because they do not model local information and the cyclical consistency of heartbeat. In this paper, we introduce a newly collected CardiacUDA dataset and a novel GraphEcho method for cardiac structure segmentation. Our GraphEcho comprises two innovative modules, the Spatial-wise Cross-domain Graph Matching (SCGM) and the Temporal Cycle Consistency (TCC) module, which utilize prior knowledge of echocardiogram videos, i.e., consistent cardiac structure across patients and centers and the heartbeat cyclical consistency, respectively. These two modules can better align global and local features from source and target domains, leading to improved UDA segmentation results. Experimental results showed that our GraphEcho outperforms existing state-of-the-art UDA segmentation methods. Our collected dataset and code will be publicly released upon acceptance. This work will lay a new and solid cornerstone for cardiac structure segmentation from echocardiogram videos. Code and dataset are available at : https://github.com/xmedlab/GraphEcho

Recently, large-scale pre-trained vision-language models have presented benefits for alleviating class imbalance in long-tailed recognition. However, the long-tailed data distribution can corrupt the representation space, where the distance between head and tail categories is much larger than the distance between two tail categories. This uneven feature space distribution causes the model to exhibit unclear and inseparable decision boundaries on the uniformly distributed test set, which lowers its performance. To address these challenges, we propose the uniformly category prototype-guided vision-language framework to effectively mitigate feature space bias caused by data imbalance. Especially, we generate a set of category prototypes uniformly distributed on a hypersphere. Category prototype-guided mechanism for image-text matching makes the features of different classes converge to these distinct and uniformly distributed category prototypes, which maintain a uniform distribution in the feature space, and improve class boundaries. Additionally, our proposed irrelevant text filtering and attribute enhancement module allows the model to ignore irrelevant noisy text and focus more on key attribute information, thereby enhancing the robustness of our framework. In the image recognition fine-tuning stage, to address the positive bias problem of the learnable classifier, we design the class feature prototype-guided classifier, which compensates for the performance of tail classes while maintaining the performance of head classes. Our method outperforms previous vision-language methods for long-tailed learning work by a large margin and achieves state-of-the-art performance.

Left-ventricular ejection fraction (LVEF) is an important indicator of heart failure. Existing methods for LVEF estimation from video require large amounts of annotated data to achieve high performance, e.g. using 10,030 labeled echocardiogram videos to achieve mean absolute error (MAE) of 4.10. Labeling these videos is time-consuming however and limits potential downstream applications to other heart diseases. This paper presents the first semi-supervised approach for LVEF prediction. Unlike general video prediction tasks, LVEF prediction is specifically related to changes in the left ventricle (LV) in echocardiogram videos. By incorporating knowledge learned from predicting LV segmentations into LVEF regression, we can provide additional context to the model for better predictions. To this end, we propose a novel Cyclical Self-Supervision (CSS) method for learning video-based LV segmentation, which is motivated by the observation that the heartbeat is a cyclical process with temporal repetition. Prediction masks from our segmentation model can then be used as additional input for LVEF regression to provide spatial context for the LV region. We also introduce teacher-student distillation to distill the information from LV segmentation masks into an end-to-end LVEF regression model that only requires video inputs. Results show our method outperforms alternative semi-supervised methods and can achieve MAE of 4.17, which is competitive with state-of-the-art supervised performance, using half the number of labels. Validation on an external dataset also shows improved generalization ability from using our method.

Surgical phase recognition is a fundamental task in computer-assisted surgery systems. Most existing works are under the supervision of expensive and time-consuming full annotations, which require the surgeons to repeat watching videos to find the precise start and end time for a surgical phase. In this paper, we introduce timestamp supervision for surgical phase recognition to train the models with timestamp annotations, where the surgeons are asked to identify only a single timestamp within the temporal boundary of a phase. This annotation can significantly reduce the manual annotation cost compared to the full annotations. To make full use of such timestamp supervisions, we propose a novel method called uncertainty-aware temporal diffusion (UATD) to generate trustworthy pseudo labels for training. Our proposed UATD is motivated by the property of surgical videos, i.e., the phases are long events consisting of consecutive frames. To be specific, UATD diffuses the single labelled timestamp to its corresponding high confident (i.e., low uncertainty) neighbour frames in an iterative way. Our study uncovers unique insights of surgical phase recognition with timestamp supervision: 1) timestamp annotation can reduce 74% annotation time compared with the full annotation, and surgeons tend to annotate those timestamps near the middle of phases; 2) extensive experiments demonstrate that our method can achieve competitive results compared with full supervision methods, while reducing manual annotation costs; 3) less is more in surgical phase recognition, i.e., less but discriminative pseudo labels outperform full but containing ambiguous frames; 4) the proposed UATD can be used as a plug-and-play method to clean ambiguous labels near boundaries between phases, and improve the performance of the current surgical phase recognition methods. Code and annotations obtained from surgeons are available at https://github.com/xmed-lab/TimeStamp-Surgical.

The understanding of language plays a key role in video grounding, where a target moment is localized according to a text query. From a biological point of view, language is naturally hierarchical, with the main clause (predicate phrase) providing coarse semantics and modifiers providing detailed descriptions. In video grounding, moments described by the main clause may exist in multiple clips of a long video, including both the ground-truth and background clips. Therefore, in order to correctly discriminate the ground-truth clip from the background ones, this co-existence leads to the negligence of the main clause, and concentrate the model on the modifiers that provide discriminative information on distinguishing the target proposal from the others. We first demonstrate this phenomenon empirically, and propose a Hierarchical Language Network (HLN) that exploits the language hierarchy, as well as a new learning approach called Multi-Instance Positive-Unlabelled Learning (MI-PUL) to alleviate the above problem. Specifically, in HLN, the localization is performed on various layers of the language hierarchy, so that the attention can be paid to different parts of the sentences, rather than only discriminative ones. Furthermore, MI-PUL allows the model to localize background clips that can be possibly described by the main clause, even without manual annotations. Therefore, the union of the two proposed components enhances the learning of the main clause, which is of critical importance in video grounding. Finally, we evaluate that our proposed HLN can plug into the current methods and improve their performance. Extensive experiments on challenging datasets show HLN significantly improve the state-of-the-art methods, especially achieving 6.15% gain in terms of [Formula: see text] on the TACoS dataset.

Automatic surgical phase recognition plays a vital role in robot-assisted surgeries. Existing methods ignored a pivotal problem that surgical phases should be classified by learning segment-level semantics instead of solely relying on frame-wise information. This paper presents a segment-attentive hierarchical consistency network (SAHC) for surgical phase recognition from videos. The key idea is to extract hierarchical high-level semantic-consistent segments and use them to refine the erroneous predictions caused by ambiguous frames. To achieve it, we design a temporal hierarchical network to generate hierarchical high-level segments. Then, we introduce a hierarchical segment-frame attention module to capture relations between the low-level frames and high-level segments. By regularizing the predictions of frames and their corresponding segments via a consistency loss, the network can generate semantic-consistent segments and then rectify the misclassified predictions caused by ambiguous low-level frames. We validate SAHC on two public surgical video datasets, i.e., the M2CAI16 challenge dataset and the Cholec80 dataset. Experimental results show that our method outperforms previous state-of-the-arts and ablation studies prove the effectiveness of our proposed modules. Our code has been released at: https://github.com/xmed-lab/SAHC.

Current approaches for video grounding propose kinds of complex architectures to capture the video-text relations, and have achieved impressive improvements. However, it is hard to learn the complicated multi-modal relations by only architecture designing in fact. In this paper, we introduce a novel Support-set Based Cross-Supervision (Sscs) module which can improve existing methods during training phase without extra inference cost. The proposed Sscs module contains two main components, i.e., discriminative contrastive objective and generative caption objective. The contrastive objective aims to learn effective representations by contrastive learning, while the caption objective can train a powerful video encoder supervised by texts. Due to the co-existence of some visual entities in both ground-truth and background intervals, i.e. mutual exclusion, naively contrastive learning is unsuitable to video grounding. We address the problem by boosting the cross-supervision with the support-set concept, which collects visual information from the whole video and eliminates the mutual exclusion of entities. Combined with the original objectives, Sscs can enhance the abilities of multi-modal relation modeling for existing approaches. We extensively evaluate Sscs on three challenging datasets, and show that our method can improve current state-of-the-art methods by large margins, especially 6.35% in terms of [email protected] on Charades-STA.

In temporal action localization (TAL), semi-supervised learning is a promising technique to mitigate the cost of precise boundary annotations. Semi-supervised approaches employing consistency regularization (CR), encouraging models to be robust to the perturbed inputs, have achieved great success in image classification problems. The success of CR is largely depended on the perturbations, where instances are perturbed to train a robust model without altering their semantic information. However, the perturbations for image or video classification tasks are not fit to apply to TAL. Since videos in TAL are too long to train the model with raw videos in an end-to-end manner. In this paper, we devise a method named K-farthest crossover to construct perturbations based on video features and apply it to TAL. Motivated by the observation that features in the same action instance become more and more similar during the training process while those in different action instances or backgrounds become more and more divergent, we add perturbations to each feature along temporal axis and adopt CR to encourage the model to retain this observation. Specifically, for a feature, we first find the top-k dissimilar features and average them to form a perturbation. Then, similar to chromosomal crossover, we select a large part of the feature and a small part of the perturbation to recombine a perturbed feature, which preserves the feature semantics yet enough discrepancy.

Weakly Supervised Temporal Action Localization (WTAL) aims to localize action segments in untrimmed videos with only video-level category labels in the training phase. In WTAL, an action generally consists of a series of sub-actions, and different categories of actions may share the common sub-actions. However, to distinguish different categories of actions with only video-level class labels, current WTAL models tend to focus on discriminative sub-actions of the action, while ignoring those common sub-actions shared with different categories of actions. This negligence of common sub-actions would lead to the located action segments incomplete, i.e., only containing discriminative sub-actions. Different from current approaches of designing complex network architectures to explore more complete actions, in this paper, we introduce a novel supervision method named multi-hierarchical category supervision (MHCS) to find more sub-actions rather than only the discriminative ones. Specifically, action categories sharing similar sub-actions will be constructed as super-classes through hierarchical clustering. Hence, training with the new generated super-classes would encourage the model to pay more attention to the common sub-actions, which are ignored training with the original classes. Furthermore, our proposed MHCS is model-agnostic and non-intrusive, which can be directly applied to existing methods without changing their structures. Through extensive experiments, we verify that our supervision method can improve the performance of four state-of-the-art WTAL methods on three public datasets: THUMOS14, ActivityNet1.2, and ActivityNet1.3.

Facial expression recognition is of significant importance in criminal investigation and digital entertainment. Under unconstrained conditions, existing expression datasets are highly class-imbalanced, and the similarity between expressions is high. Previous methods tend to improve the performance of facial expression recognition through deeper or wider network structures, resulting in increased storage and computing costs. In this paper, we propose a new adaptive supervised objective named AdaReg loss, re-weighting category importance coefficients to address this class imbalance and increasing the discrimination power of expression representations. Inspired by human beings' cognitive mode, an innovative coarse-fine (C-F) labels strategy is designed to guide the model from easy to difficult to classify highly similar representations. On this basis, we propose a novel training framework named the emotional education mechanism (EEM) to transfer knowledge, composed of a knowledgeable teacher network (KTN) and a self-taught student network (STSN). Specifically, KTN integrates the outputs of coarse and fine streams, learning expression representations from easy to difficult. Under the supervision of the pre-trained KTN and existing learning experience, STSN can maximize the potential performance and compress the original KTN. Extensive experiments on public benchmarks demonstrate that the proposed method achieves superior performance compared to current state-of-the-art frameworks with 88.07% on RAF-DB, 63.97% on AffectNet and 90.49% on FERPlus.

In capsule networks (CapsNets), the capsule is made up of collections of neurons. Their adjacent capsule layers are connected using routing-by-agreement mechanisms in an unsupervised way. The routing-by-agreement mechanisms have two main drawbacks: a) too many parameters and high computation complexity; b) the cluster distribution assumptions of these routing mechanisms may not hold in some complex real-world data. In this paper, we propose a novel Group Feedback Capsule Network (GF-CapsNet) which adopts a supervised routing strategy called group-routing. Compared with the previous routing strategies which globally transform each capsule, Group-routing equally splits capsules into groups where capsules locally share the same transformation weights, reducing routing parameters. To address the second drawback, we devise a distance network to directly predict capsules in a supervised way without making distribution assumptions. Our proposed group-routing captures local information of low-level capsules by group-wise transformation and supervisedly predicts high-level ones in a feedback way to address two drawbacks respectively. We conduct experiments on CIFAR-10/100 and SVHN datasets and the results show that our method can perform better against state-of-the-arts.

In capsule networks, the mapping of low-level capsules to high-level capsules is achieved by a routing-by-agreement algorithm. Since the capsule is made up of collections of neurons and the routing mechanism involves all the capsules instead of simply discarding some of the neurons like Max-Pooling, the capsule network has stronger representation ability than the traditional neural network. However, considering too much low-level capsules' information will cause its corresponding upper layer capsules to be interfered by other irrelevant information or noise capsules. Therefore, the original capsule network does not perform well on complex data structure. What's worse, computational complexity becomes a bottleneck in dealing with large data networks. In order to solve these shortcomings, this paper proposes a group reconstruction and max-pooling residual capsule network (GRMR-CapsNet). We build a block in which all capsules are divided into different groups and perform group reconstruction routing algorithm to obtain the corresponding high-level capsules. Between the lower and higher layers, Capsule Max-Pooling is adopted to prevent overfitting. We conduct experiments on CIFAR-10/100 and SVHN datasets and the results show that our method can perform better against state-of-the-arts.