This work focuses on multi-shot semi-supervised video object segmentation (MVOS), which aims at segmenting the target object indicated by an initial mask throughout a video with multiple shots. While existing VOS methods mainly focus on single-shot videos, they often fail to handle shot discontinuities, thereby limiting their real-world applicability. Furthermore, the lack of annotated multi-shot data poses a major challenge for MVOS research. To address these issues, we propose a transition mimicking data augmentation strategy (TMA) that enables cross-shot generalization using single-shot data, and a transition-aware method, Segment Anything Across Shots (SAAS), which detects and comprehends shot transitions during inference. To support evaluation and future study in MVOS, we introduce Cut-VOS, a new MVOS benchmark with dense mask annotations, diverse object categories, and high-frequency transitions. Extensive experiments on YouMVOS and Cut-VOS demonstrate that the proposed SAAS achieves state-of-the-art performance by effectively mimicking, understanding, and segmenting across complex transitions.

Referring audio-visual segmentation (RAVS) has recently seen significant advancements, yet challenges remain in integrating multimodal information and deeply understanding and reasoning about audiovisual content. To extend the boundaries of RAVS and facilitate future research in this field, we propose Omnimodal Referring Audio-Visual Segmentation (OmniAVS), a new dataset containing 2,104 videos and 61,095 multimodal referring expressions. OmniAVS stands out with three key innovations: (1) 8 types of multimodal expressions that flexibly combine text, speech, sound, and visual cues; (2) an emphasis on understanding audio content beyond just detecting their presence; and (3) the inclusion of complex reasoning and world knowledge in expressions. Furthermore, we introduce Omnimodal Instructed Segmentation Assistant (OISA), to address the challenges of multimodal reasoning and fine-grained understanding of audiovisual content in OmniAVS. OISA uses MLLM to comprehend complex cues and perform reasoning-based segmentation. Extensive experiments show that OISA outperforms existing methods on OmniAVS and achieves competitive results on other related tasks.

This paper proposes a large-scale multi-modal dataset for referring motion expression video segmentation, focusing on segmenting and tracking target objects in videos based on language description of objects' motions. Existing referring video segmentation datasets often focus on salient objects and use language expressions rich in static attributes, potentially allowing the target object to be identified in a single frame. Such datasets underemphasize the role of motion in both videos and languages. To explore the feasibility of using motion expressions and motion reasoning clues for pixel-level video understanding, we introduce MeViS, a dataset containing 33,072 human-annotated motion expressions in both text and audio, covering 8,171 objects in 2,006 videos of complex scenarios. We benchmark 15 existing methods across 4 tasks supported by MeViS, including 6 referring video object segmentation (RVOS) methods, 3 audio-guided video object segmentation (AVOS) methods, 2 referring multi-object tracking (RMOT) methods, and 4 video captioning methods for the newly introduced referring motion expression generation (RMEG) task. The results demonstrate weaknesses and limitations of existing methods in addressing motion expression-guided video understanding. We further analyze the challenges and propose an approach LMPM++ for RVOS/AVOS/RMOT that achieves new state-of-the-art results. Our dataset provides a platform that facilitates the development of motion expression-guided video understanding algorithms in complex video scenes.

Large Vision-Language Models (LVLMs) show significant strides in general-propose multimodal applications such as visual dialogue and embodied navigation. However, existing multimodal evaluation benchmarks cover a limited number of multimodal tasks testing rudimentary capabilities, falling short in tracking LVLM development. In this study, we present MMT-Bench, a comprehensive benchmark designed to assess LVLMs across massive multimodal tasks requiring expert knowledge and deliberate visual recognition, localization, and reasoning. MMT-Bench comprises $31,325$ meticulously curated multi-choice visual questions from various multimodal scenarios such as vehicle driving and embodied navigation, covering $32$ core meta-tasks and $162$ subtasks in multimodal understanding. Due to its extensive task coverage, MMT-Bench enables the evaluation of LVLMs using a task map, facilitating the discovery of in- and out-of-domain tasks. Evaluation results involving $20$ publicly available LVLMs such as the proprietary GeminiProVision model, underscore the significant challenges posed by MMT-Bench. We anticipate that MMT-Bench will inspire the community to develop next-generation multimodal foundation models aimed at achieving general-purpose multimodal intelligence.

Multi-turn visual conversation is an important ability of real-world AI assistants. However, the related evaluation benchmark is missed. This paper presents ConvBench, a multi-turn conversation benchmark with hierarchical capabilities ablation evaluation for Large Vision-Language Models (LVLMs). ConvBench comprises 577 curated multi-turn conversations, encompassing 215 tasks. These tasks are broad and open-ended, which resemble real-world user behaviors. ConvBench progressively examines the LVLMs' perception, reasoning, and creativity capabilities in each conversation and can decouple these capabilities in evaluations and thus perform reliable error attribution. Besides, considering the diversity of open-ended questions, we introduce an efficient and reliable automatic evaluation framework. Experimental results reveal that ConvBench is a significant challenge for current LVLMs, even for GPT4V, which achieves only a 39.51% score. Besides, we have some insightful findings, such as the weak perception of LVLMs inhibits authentic strengths in reasoning and creation. We believe our design of hierarchical capabilities, decoupling capabilities evaluation, and multi-turn conversation can blaze a new trail in LVLMs evaluation. Code and benchmark are released at https://github.com/shirlyliu64/ConvBench.

Named entity disambiguation (NED) finds the specific meaning of an entity mention in a particular context and links it to a target entity. With the emergence of multimedia, the modalities of content on the Internet have become more diverse, which poses difficulties for traditional NED, and the vast amounts of information make it impossible to manually label every kind of ambiguous data to train a practical NED model. In response to this situation, we present MMGraph, which uses multimodal graph convolution to aggregate visual and contextual language information for accurate entity disambiguation for short texts, and a self-supervised simple triplet network (SimTri) that can learn useful representations in multimodal unlabeled data to enhance the effectiveness of NED models. We evaluated these approaches on a new dataset, MMFi, which contains multimodal supervised data and large amounts of unlabeled data. Our experiments confirm the state-of-the-art performance of MMGraph on two widely used benchmarks and MMFi. SimTri further improves the performance of NED methods. The dataset and code are available at https://github.com/LanceZPF/NNED_MMGraph.

The discrimination of instance embeddings plays a vital role in associating instances across time for online video instance segmentation (VIS). Instance embedding learning is directly supervised by the contrastive loss computed upon the contrastive items (CIs), which are sets of anchor/positive/negative embeddings. Recent online VIS methods leverage CIs sourced from one reference frame only, which we argue is insufficient for learning highly discriminative embeddings. Intuitively, a possible strategy to enhance CIs is replicating the inference phase during training. To this end, we propose a simple yet effective training strategy, called Consistent Training for Online VIS (CTVIS), which devotes to aligning the training and inference pipelines in terms of building CIs. Specifically, CTVIS constructs CIs by referring inference the momentum-averaged embedding and the memory bank storage mechanisms, and adding noise to the relevant embeddings. Such an extension allows a reliable comparison between embeddings of current instances and the stable representations of historical instances, thereby conferring an advantage in modeling VIS challenges such as occlusion, re-identification, and deformation. Empirically, CTVIS outstrips the SOTA VIS models by up to +5.0 points on three VIS benchmarks, including YTVIS19 (55.1% AP), YTVIS21 (50.1% AP) and OVIS (35.5% AP). Furthermore, we find that pseudo-videos transformed from images can train robust models surpassing fully-supervised ones.

Compared with the progress made on human activity classification, much less success has been achieved on human interaction understanding (HIU). Apart from the latter task is much more challenging, the main causation is that recent approaches learn human interactive relations via shallow graphical representations, which are inadequate to model complicated human interactive-relations. This paper proposes a deep consistency-aware framework aiming at tackling the grouping and labelling inconsistencies in HIU. This framework consists of three components, including a backbone CNN to extract image features, a factor graph network to implicitly learn higher-order consistencies among labelling and grouping variables, and a consistency-aware reasoning module to explicitly enforcing consistencies. The last module is inspired by our key observation that the consistency-aware reasoning bias can be embedded into an energy function or a particular loss function, minimizing which delivers consistent predictions. An efficient mean-field inference algorithm is proposed, such that all modules of our network could be trained in an end-to-end fashion. Experimental results demonstrate that the two proposed consistency-learning modules complement each other, and both make considerable contributions in achieving leading performance on three benchmarks of HIU. The effectiveness of the proposed approach is further validated by experiments on detecting human-object interactions.

Most instance segmentation models are not end-to-end trainable due to either the incorporation of proposal estimation (RPN) as a pre-processing or non-maximum suppression (NMS) as a post-processing. Here we propose a novel end-to-end instance segmentation method termed ISDA. It reshapes the task into predicting a set of object masks, which are generated via traditional convolution operation with learned position-aware kernels and features of objects. Such kernels and features are learned by leveraging a deformable attention network with multi-scale representation. Thanks to the introduced set-prediction mechanism, the proposed method is NMS-free. Empirically, ISDA outperforms Mask R-CNN (the strong baseline) by 2.6 points on MS-COCO, and achieves leading performance compared with recent models. Code will be available soon.

With the increasing health concerns on diet, it's worthwhile to develop an intelligent assistant that can help users eat healthier. Such an assistant can automatically give personal advice for the user's diet and generate health report about eating on a regular basis. To boost the research on such diet assistant, we establish a real-world foodlog database using various methods such as filter, cluster and graph convolutional network. This database is built based on real-world lifelog and medical data, which is named as Real-World Multimodal Foodlog (RWMF). It contains 7500 multimodal pairs, and each pair consists of a food image paired with a line of personal biometrics data (such as Blood Glucose) and a textual food description of food composition paired with a line of food nutrition data. In this paper, we present the detailed procedures for setting up the database. We evaluate the performance of RWMF using different food image classification and cross-modal retrieval approaches. We also test the performance of multimodal fusion on RWMF through ablation experiments. The experimental results show that the RWMF database is quite challenging and can be widely used to evaluate the performance of food analysis methods based on multimodal data.