Basura Fernando

Debaditya Roy, Basura Fernando

The problem of anticipating human actions is an inherently uncertain one. However, we can reduce this uncertainty if we have a sense of the goal that the actor is trying to achieve. Here, we present an action anticipation model that leverages goal information for the purpose of reducing the uncertainty in future predictions. Since we do not possess goal information or the observed actions during inference, we resort to visual representation to encapsulate information about both actions and goals. Through this, we derive a novel concept called abstract goal which is conditioned on observed sequences of visual features for action anticipation. We design the abstract goal as a distribution whose parameters are estimated using a variational recurrent network. We sample multiple candidates for the next action and introduce a goal consistency measure to determine the best candidate that follows from the abstract goal. Our method obtains impressive results on the very challenging Epic-Kitchens55 (EK55), EK100, and EGTEA Gaze+ datasets. We obtain absolute improvements of +13.69, +11.24, and +5.19 for Top-1 verb, Top-1 noun, and Top-1 action anticipation accuracy respectively over prior state-of-the-art methods for seen kitchens (S1) of EK55. Similarly, we also obtain significant improvements in the unseen kitchens (S2) set for Top-1 verb (+10.75), noun (+5.84) and action (+2.87) anticipation. Similar trend is observed for EGTEA Gaze+ dataset, where absolute improvement of +9.9, +13.1 and +6.8 is obtained for noun, verb, and action anticipation. It is through the submission of this paper that our method is currently the new state-of-the-art for action anticipation in EK55 and EGTEA Gaze+ https://competitions.codalab.org/competitions/20071#results Code available at https://github.com/debadityaroy/Abstract_Goal

Hao Zhang, Yeo Keat Ee, Basura Fernando

Visual abductive reasoning aims to make likely explanations for visual observations. We propose a simple yet effective Region Conditioned Adaptation, a hybrid parameter-efficient fine-tuning method that equips the frozen CLIP with the ability to infer explanations from local visual cues. We encode "local hints" and "global contexts" into visual prompts of the CLIP model separately at fine and coarse-grained levels. Adapters are used for fine-tuning CLIP models for downstream tasks and we design a new attention adapter, that directly steers the focus of the attention map with trainable query and key projections of a frozen CLIP model. Finally, we train our new model with a modified contrastive loss to regress the visual feature simultaneously toward features of literal description and plausible explanations. The loss enables CLIP to maintain both perception and reasoning abilities. Experiments on the Sherlock visual abductive reasoning benchmark show that the RCA significantly outstands previous SOTAs, ranking the 1st on the leaderboards (e.g., Human Acc: RCA 31.74 $\textit{vs}$ CPT-CLIP 29.58, higher =better). We also validate the RCA is generalizable to local perception benchmarks like RefCOCO. We open-source our project at https://github.com/LUNAProject22/RPA.

Clement Tan, Chai Kiat Yeo, Cheston Tan et al.

Abductive reasoning aims to make the most likely inference for a given set of incomplete observations. In this paper, we introduce "Abductive Past Action Inference", a novel research task aimed at identifying the past actions performed by individuals within homes to reach specific states captured in a single image, using abductive inference. The research explores three key abductive inference problems: past action set prediction, past action sequence prediction, and abductive past action verification. We introduce several models tailored for abductive past action inference, including a relational graph neural network, a relational bilinear pooling model, and a relational transformer model. Notably, the newly proposed object-relational bilinear graph encoder-decoder (BiGED) model emerges as the most effective among all methods evaluated, demonstrating good proficiency in handling the intricacies of the Action Genome dataset. The contributions of this research significantly advance the ability of deep learning models to reason about current scene evidence and make highly plausible inferences about past human actions. This advancement enables a deeper understanding of events and behaviors, which can enhance decision-making and improve system capabilities across various real-world applications such as Human-Robot Interaction and Elderly Care and Health Monitoring. Code and data available at https://github.com/LUNAProject22/AAR

Debaditya Roy, Ramanathan Rajendiran, Basura Fernando

Human-object interaction is one of the most important visual cues and we propose a novel way to represent human-object interactions for egocentric action anticipation. We propose a novel transformer variant to model interactions by computing the change in the appearance of objects and human hands due to the execution of the actions and use those changes to refine the video representation. Specifically, we model interactions between hands and objects using Spatial Cross-Attention (SCA) and further infuse contextual information using Trajectory Cross-Attention to obtain environment-refined interaction tokens. Using these tokens, we construct an interaction-centric video representation for action anticipation. We term our model InAViT which achieves state-of-the-art action anticipation performance on large-scale egocentric datasets EPICKTICHENS100 (EK100) and EGTEA Gaze+. InAViT outperforms other visual transformer-based methods including object-centric video representation. On the EK100 evaluation server, InAViT is the top-performing method on the public leaderboard (at the time of submission) where it outperforms the second-best model by 3.3% on mean-top5 recall.

Debaditya Roy, Dhruv Verma, Basura Fernando

Situation Recognition is the task of generating a structured summary of what is happening in an image using an activity verb and the semantic roles played by actors and objects. In this task, the same activity verb can describe a diverse set of situations as well as the same actor or object category can play a diverse set of semantic roles depending on the situation depicted in the image. Hence a situation recognition model needs to understand the context of the image and the visual-linguistic meaning of semantic roles. Therefore, we leverage the CLIP foundational model that has learned the context of images via language descriptions. We show that deeper-and-wider multi-layer perceptron (MLP) blocks obtain noteworthy results for the situation recognition task by using CLIP image and text embedding features and it even outperforms the state-of-the-art CoFormer, a Transformer-based model, thanks to the external implicit visual-linguistic knowledge encapsulated by CLIP and the expressive power of modern MLP block designs. Motivated by this, we design a cross-attention-based Transformer using CLIP visual tokens that model the relation between textual roles and visual entities. Our cross-attention-based Transformer known as ClipSitu XTF outperforms existing state-of-the-art by a large margin of 14.1\% on semantic role labelling (value) for top-1 accuracy using imSitu dataset. {Similarly, our ClipSitu XTF obtains state-of-the-art situation localization performance.} We will make the code publicly available.

Ishaan Singh Rawal, Alexander Matyasko, Shantanu Jaiswal et al.

While VideoQA Transformer models demonstrate competitive performance on standard benchmarks, the reasons behind their success are not fully understood. Do these models capture the rich multimodal structures and dynamics from video and text jointly? Or are they achieving high scores by exploiting biases and spurious features? Hence, to provide insights, we design $\textit{QUAG}$ (QUadrant AveraGe), a lightweight and non-parametric probe, to conduct dataset-model combined representation analysis by impairing modality fusion. We find that the models achieve high performance on many datasets without leveraging multimodal representations. To validate QUAG further, we design $\textit{QUAG-attention}$, a less-expressive replacement of self-attention with restricted token interactions. Models with QUAG-attention achieve similar performance with significantly fewer multiplication operations without any finetuning. Our findings raise doubts about the current models' abilities to learn highly-coupled multimodal representations. Hence, we design the $\textit{CLAVI}$ (Complements in LAnguage and VIdeo) dataset, a stress-test dataset curated by augmenting real-world videos to have high modality coupling. Consistent with the findings of QUAG, we find that most of the models achieve near-trivial performance on CLAVI. This reasserts the limitations of current models for learning highly-coupled multimodal representations, that is not evaluated by the current datasets (project page: https://dissect-videoqa.github.io ).

Kian Boon Koh, Basura Fernando

Collection of real world annotations for training semantic segmentation models is an expensive process. Unsupervised domain adaptation (UDA) tries to solve this problem by studying how more accessible data such as synthetic data can be used to train and adapt models to real world images without requiring their annotations. Recent UDA methods applies self-learning by training on pixel-wise classification loss using a student and teacher network. In this paper, we propose the addition of a consistency regularization term to semi-supervised UDA by modelling the inter-pixel relationship between elements in networks' output. We demonstrate the effectiveness of the proposed consistency regularization term by applying it to the state-of-the-art DAFormer framework and improving mIoU19 performance on the GTA5 to Cityscapes benchmark by 0.8 and mIou16 performance on the SYNTHIA to Cityscapes benchmark by 1.2.

Arushi Goel, Basura Fernando, Frank Keller et al.

Coreference resolution aims to identify words and phrases which refer to same entity in a text, a core task in natural language processing. In this paper, we extend this task to resolving coreferences in long-form narrations of visual scenes. First we introduce a new dataset with annotated coreference chains and their bounding boxes, as most existing image-text datasets only contain short sentences without coreferring expressions or labeled chains. We propose a new technique that learns to identify coreference chains using weak supervision, only from image-text pairs and a regularization using prior linguistic knowledge. Our model yields large performance gains over several strong baselines in resolving coreferences. We also show that coreference resolution helps improving grounding narratives in images.

Yeo Keat Ee, Debaditya Roy, Chen Li et al.

Temporal action segmentation (TAS) divides untrimmed videos into labeled action segments. While fully supervised methods have advanced the field, challenges such as action variability, ambiguous boundaries, and high annotation costs remain, especially in new or low-resource domains. Grammar-based approaches improve segmentation with structural priors but rely on complex parsing limiting scalability. In this work, we propose a lightweight, constraint-based refinement framework that enhances TAS predictions by integrating statistical structural priors such as transition confidence, action boundary sets, and per-class duration, that can be directly extracted from annotated data. These constraints are integrated into a modified Viterbi decoding algorithm, allowing inference-time refinement without retraining or added model complexity. Our approach improves both fully and semi-supervised TAS models by correcting structural prediction errors while maintaining high efficiency. Code is available at https://github.com/LUNAProject22/CAD

Xinyu Zhang, Yuxuan Dong, Yanrui Wu et al.

Large language models demonstrate remarkable capabilities across various domains, especially mathematics and logic reasoning. However, current evaluations overlook physics-based reasoning - a complex task requiring physics theorems and constraints. We present PhysReason, a 1,200-problem benchmark comprising knowledge-based (25%) and reasoning-based (75%) problems, where the latter are divided into three difficulty levels (easy, medium, hard). Notably, problems require an average of 8.1 solution steps, with hard requiring 15.6, reflecting the complexity of physics-based reasoning. We propose the Physics Solution Auto Scoring Framework, incorporating efficient answer-level and comprehensive step-level evaluations. Top-performing models like Deepseek-R1, Gemini-2.0-Flash-Thinking, and o3-mini-high achieve less than 60% on answer-level evaluation, with performance dropping from knowledge questions (75.11%) to hard problems (31.95%). Through step-level evaluation, we identified four key bottlenecks: Physics Theorem Application, Physics Process Understanding, Calculation, and Physics Condition Analysis. These findings position PhysReason as a novel and comprehensive benchmark for evaluating physics-based reasoning capabilities in large language models. Our code and data will be published at https:/dxzxy12138.github.io/PhysReason.

Vincent Tao Hu, Wenzhe Yin, Pingchuan Ma et al.

Human motion synthesis is a fundamental task in computer animation. Recent methods based on diffusion models or GPT structure demonstrate commendable performance but exhibit drawbacks in terms of slow sampling speeds and error accumulation. In this paper, we propose \emph{Motion Flow Matching}, a novel generative model designed for human motion generation featuring efficient sampling and effectiveness in motion editing applications. Our method reduces the sampling complexity from thousand steps in previous diffusion models to just ten steps, while achieving comparable performance in text-to-motion and action-to-motion generation benchmarks. Noticeably, our approach establishes a new state-of-the-art Fréchet Inception Distance on the KIT-ML dataset. What is more, we tailor a straightforward motion editing paradigm named \emph{sampling trajectory rewriting} leveraging the ODE-style generative models and apply it to various editing scenarios including motion prediction, motion in-between prediction, motion interpolation, and upper-body editing. Our code will be released.

Dhruv Verma, Debaditya Roy, Basura Fernando

Situation recognition refers to the ability of an agent to identify and understand various situations or contexts based on available information and sensory inputs. It involves the cognitive process of interpreting data from the environment to determine what is happening, what factors are involved, and what actions caused those situations. This interpretation of situations is formulated as a semantic role labeling problem in computer vision-based situation recognition. Situations depicted in images and videos hold pivotal information, essential for various applications like image and video captioning, multimedia retrieval, autonomous systems and event monitoring. However, existing methods often struggle with ambiguity and lack of context in generating meaningful and accurate predictions. Leveraging multimodal models such as CLIP, we propose ClipSitu, which sidesteps the need for full fine-tuning and achieves state-of-the-art results in situation recognition and localization tasks. ClipSitu harnesses CLIP-based image, verb, and role embeddings to predict nouns fulfilling all the roles associated with a verb, providing a comprehensive understanding of depicted scenarios. Through a cross-attention Transformer, ClipSitu XTF enhances the connection between semantic role queries and visual token representations, leading to superior performance in situation recognition. We also propose a verb-wise role prediction model with near-perfect accuracy to create an end-to-end framework for producing situational summaries for out-of-domain images. We show that situational summaries empower our ClipSitu models to produce structured descriptions with reduced ambiguity compared to generic captions. Finally, we extend ClipSitu to video situation recognition to showcase its versatility and produce comparable performance to state-of-the-art methods.

Kenneth Kwok, Basura Fernando, Qianli Xu et al.

This paper addresses the topic of robustness under sensing noise, ambiguous instructions, and human-robot interaction. We take a radically different tack to the issue of reliable embodied AI: instead of focusing on formal verification methods aimed at achieving model predictability and robustness, we emphasise the dynamic, ambiguous and subjective nature of human-robot interactions that requires embodied AI systems to perceive, interpret, and respond to human intentions in a manner that is consistent, comprehensible and aligned with human expectations. We argue that when embodied agents operate in human environments that are inherently social, multimodal, and fluid, reliability is contextually determined and only has meaning in relation to the goals and expectations of humans involved in the interaction. This calls for a fundamentally different approach to achieving reliable embodied AI that is centred on building and updating an accessible "explicit world model" representing the common ground between human and AI, that is used to align robot behaviours with human expectations.

Arushi Goel, Basura Fernando, Frank Keller et al.

In this paper, we study multimodal coreference resolution, specifically where a longer descriptive text, i.e., a narration is paired with an image. This poses significant challenges due to fine-grained image-text alignment, inherent ambiguity present in narrative language, and unavailability of large annotated training sets. To tackle these challenges, we present a data efficient semi-supervised approach that utilizes image-narration pairs to resolve coreferences and narrative grounding in a multimodal context. Our approach incorporates losses for both labeled and unlabeled data within a cross-modal framework. Our evaluation shows that the proposed approach outperforms strong baselines both quantitatively and qualitatively, for the tasks of coreference resolution and narrative grounding.

Binqian Xu, Xiangbo Shu, Haiyang Mei et al.

Multimodal Large Language Models (MLLMs) have made significant advancements, demonstrating powerful capabilities in processing and understanding multimodal data. Fine-tuning MLLMs with Federated Learning (FL) allows for expanding the training data scope by including private data sources, thereby enhancing their practical applicability in privacy-sensitive domains. However, current research remains in the early stage, particularly in addressing the \textbf{multimodal heterogeneities} in real-world applications. In this paper, we introduce a benchmark to evaluate the performance of federated fine-tuning of MLLMs across various multimodal heterogeneous scenarios, laying the groundwork for future research in the field. Our benchmark includes two lightweight MLLMs, two downstream tasks, three evaluation metrics, and five datasets across three domains, along with six comparison baselines, covering over ten types of modality heterogeneities across four multimodal scenarios. To address the challenges posed by multimodal heterogeneity, we develop a general FedMLLM framework that integrates classic FL methods alongside two modality-agnostic strategies. Extensive experimental results show that our proposed FL paradigm improves the performance of MLLMs by broadening the range of training data and mitigating multimodal heterogeneity. Code is available in supplementary materials.

Thanh-Son Nguyen, Hong Yang, Tzeh Yuan Neoh et al.

We introduce PKR-QA (Procedural Knowledge Reasoning Question Answering), a new benchmark for question answering over procedural tasks that require structured reasoning. PKR-QA is constructed semi-automatically using a procedural knowledge graph (PKG), which encodes task-specific knowledge across diverse domains. The PKG is built by curating and linking information from the COIN instructional video dataset and the ontology, enriched with commonsense knowledge from ConceptNet and structured outputs from Large Language Models (LLMs), followed by manual verification. To generate question-answer pairs, we design graph traversal templates where each template is applied systematically over PKG. To enable interpretable reasoning, we propose a neurosymbolic approach called Knowledge Module Learning (KML), which learns procedural relations via neural modules and composes them for structured reasoning with LLMs. Experiments demonstrate that this paradigm improves reasoning performance on PKR-QA and enables step-by-step reasoning traces that facilitate interpretability. Code and dataset will be released soon https://github.com/LUNAProject22/KML.

Chinthani Sugandhika, Chen Li, Deepu Rajan et al.

Spatio-temporal scene graph generation (ST-SGG) aims to model objects and their evolving relationships across video frames, enabling interpretable representations for downstream reasoning tasks such as video captioning and visual question answering. Despite recent advancements in DETR-style single-stage ST-SGG models, they still suffer from several key limitations. First, while these models rely on attention-based learnable queries as a core component, these learnable queries are semantically uninformed and instance-agnostically initialized. Second, these models rely exclusively on unimodal visual features for predicate classification. To address these challenges, we propose VOST-SGG, a VLM-aided one-stage ST-SGG framework that integrates the common sense reasoning capabilities of vision-language models (VLMs) into the ST-SGG pipeline. First, we introduce the dual-source query initialization strategy that disentangles what to attend to from where to attend, enabling semantically grounded what-where reasoning. Furthermore, we propose a multi-modal feature bank that fuses visual, textual, and spatial cues derived from VLMs for improved predicate classification. Extensive experiments on the Action Genome dataset demonstrate that our approach achieves state-of-the-art performance, validating the effectiveness of integrating VLM-aided semantic priors and multi-modal features for ST-SGG. We will release the code at https://github.com/LUNAProject22/VOST.

Chinthani Sugandhika, Chen Li, Deepu Rajan et al.

Large Video-Language Models (Video-LMs) have achieved impressive progress in multimodal understanding, yet their reasoning remains weakly grounded in space and time. We present Know-Show, a new benchmark designed to evaluate spatio-temporal grounded reasoning, the ability of a model to reason about actions and their semantics while simultaneously grounding its inferences in visual and temporal evidence. Know-Show unifies reasoning and localization within a single evaluation framework consisting of five complementary scenarios across spatial (person, object, person-object, and hand-object) and temporal dimensions. Built from Charades, Action Genome, and Ego4D with 2.5K human-authored questions, the benchmark exposes significant gaps between current Video-LMs and human reasoning. To bridge this gap, we propose GRAM, a training-free plug-in that augments Video-LMs with fine-grained grounding through attention-based video token selection and explicit timestamp encoding. Extensive experiments across open and closed Video-LMs (Qwen, VideoLLaVA, GPT-4o, and Gemini, etc.) reveal that existing models struggle to "show what they know" and vice versa, especially in fine-grained hand-object interactions. Know-Show establishes a unified standard for assessing grounded reasoning in video-language understanding and provides insights toward developing interpretable and reliable multimodal reasoning systems. We will release the dataset and the code at https://github.com/LUNAProject22/Know-Show.

Chen Li, Chinthani Sugandhika, Yeo Keat Ee et al.

Existing human motion Q\&A methods rely on explicit program execution, where the requirement for manually defined functional modules may limit the scalability and adaptability. To overcome this, we propose an implicit program-guided motion reasoning (IMoRe) framework that unifies reasoning across multiple query types without manually designed modules. Unlike existing implicit reasoning approaches that infer reasoning operations from question words, our model directly conditions on structured program functions, ensuring a more precise execution of reasoning steps. Additionally, we introduce a program-guided reading mechanism, which dynamically selects multi-level motion representations from a pretrained motion Vision Transformer (ViT), capturing both high-level semantics and fine-grained motion cues. The reasoning module iteratively refines memory representations, leveraging structured program functions to extract relevant information for different query types. Our model achieves state-of-the-art performance on Babel-QA and generalizes to a newly constructed motion Q\&A dataset based on HuMMan, demonstrating its adaptability across different motion reasoning datasets. Code and dataset are available at: https://github.com/LUNAProject22/IMoRe.

Shantanu Jaiswal, Basura Fernando, Cheston Tan

Attention modules for Convolutional Neural Networks (CNNs) are an effective method to enhance performance on multiple computer-vision tasks. While existing methods appropriately model channel-, spatial- and self-attention, they primarily operate in a feedforward bottom-up manner. Consequently, the attention mechanism strongly depends on the local information of a single input feature map and does not incorporate relatively semantically-richer contextual information available at higher layers that can specify "what and where to look" in lower-level feature maps through top-down information flow. Accordingly, in this work, we propose a lightweight top-down attention module (TDAM) that iteratively generates a "visual searchlight" to perform channel and spatial modulation of its inputs and outputs more contextually-relevant feature maps at each computation step. Our experiments indicate that TDAM enhances the performance of CNNs across multiple object-recognition benchmarks and outperforms prominent attention modules while being more parameter and memory efficient. Further, TDAM-based models learn to "shift attention" by localizing individual objects or features at each computation step without any explicit supervision resulting in a 5% improvement for ResNet50 on weakly-supervised object localization. Source code and models are publicly available at: https://github.com/shantanuj/TDAM_Top_down_attention_module .

Paritosh Parmar, Eric Peh, Ruirui Chen et al.

Causal video question answering (QA) has garnered increasing interest, yet existing datasets often lack depth in causal reasoning. To address this gap, we capitalize on the unique properties of cartoons and construct CausalChaos!, a novel, challenging causal Why-QA dataset built upon the iconic "Tom and Jerry" cartoon series. Cartoons use the principles of animation that allow animators to create expressive, unambiguous causal relationships between events to form a coherent storyline. Utilizing these properties, along with thought-provoking questions and multi-level answers (answer and detailed causal explanation), our questions involve causal chains that interconnect multiple dynamic interactions between characters and visual scenes. These factors demand models to solve more challenging, yet well-defined causal relationships. We also introduce hard incorrect answer mining, including a causally confusing version that is even more challenging. While models perform well, there is much room for improvement, especially, on open-ended answers. We identify more advanced/explicit causal relationship modeling & joint modeling of vision and language as the immediate areas for future efforts to focus upon. Along with the other complementary datasets, our new challenging dataset will pave the way for these developments in the field.

Shantanu Jaiswal, Debaditya Roy, Basura Fernando et al.

Complex visual reasoning and question answering (VQA) is a challenging task that requires compositional multi-step processing and higher-level reasoning capabilities beyond the immediate recognition and localization of objects and events. Here, we introduce a fully neural Iterative and Parallel Reasoning Mechanism (IPRM) that combines two distinct forms of computation -- iterative and parallel -- to better address complex VQA scenarios. Specifically, IPRM's "iterative" computation facilitates compositional step-by-step reasoning for scenarios wherein individual operations need to be computed, stored, and recalled dynamically (e.g. when computing the query "determine the color of pen to the left of the child in red t-shirt sitting at the white table"). Meanwhile, its "parallel" computation allows for the simultaneous exploration of different reasoning paths and benefits more robust and efficient execution of operations that are mutually independent (e.g. when counting individual colors for the query: "determine the maximum occurring color amongst all t-shirts"). We design IPRM as a lightweight and fully-differentiable neural module that can be conveniently applied to both transformer and non-transformer vision-language backbones. It notably outperforms prior task-specific methods and transformer-based attention modules across various image and video VQA benchmarks testing distinct complex reasoning capabilities such as compositional spatiotemporal reasoning (AGQA), situational reasoning (STAR), multi-hop reasoning generalization (CLEVR-Humans) and causal event linking (CLEVRER-Humans). Further, IPRM's internal computations can be visualized across reasoning steps, aiding interpretability and diagnosis of its errors.

Chinthani Sugandhika, Chen Li, Deepu Rajan et al.

Graph based representation has been widely used in modelling spatio-temporal relationships in video understanding. Although effective, existing graph-based approaches focus on capturing the human-object relationships while ignoring fine-grained semantic properties of the action components. These semantic properties are crucial for understanding the current situation, such as where does the action takes place, what tools are used and functional properties of the objects. In this work, we propose a graph-based representation called Situational Scene Graph (SSG) to encode both human-object relationships and the corresponding semantic properties. The semantic details are represented as predefined roles and values inspired by situation frame, which is originally designed to represent a single action. Based on our proposed representation, we introduce the task of situational scene graph generation and propose a multi-stage pipeline Interactive and Complementary Network (InComNet) to address the task. Given that the existing datasets are not applicable to the task, we further introduce a SSG dataset whose annotations consist of semantic role-value frames for human, objects and verb predicates of human-object relations. Finally, we demonstrate the effectiveness of our proposed SSG representation by testing on different downstream tasks. Experimental results show that the unified representation can not only benefit predicate classification and semantic role-value classification, but also benefit reasoning tasks on human-centric situation understanding. We will release the code and the dataset soon.

Zhengwei Yang, Yuke Li, Qiang Sun et al.

Most existing studies on few-shot learning focus on unimodal settings, where models are trained to generalize to unseen data using a limited amount of labeled examples from a single modality. However, real-world data are inherently multi-modal, and such unimodal approaches limit the practical applications of few-shot learning. To bridge this gap, this paper introduces the Cross-modal Few-Shot Learning (CFSL) task, which aims to recognize instances across multiple modalities while relying on scarce labeled data. This task presents unique challenges compared to classical few-shot learning arising from the distinct visual attributes and structural disparities inherent to each modality. To tackle these challenges, we propose a Generative Transfer Learning (GTL) framework by simulating how humans abstract and generalize concepts. Specifically, the GTL jointly estimates the latent shared concept across modalities and the in-modality disturbance through a generative structure. Establishing the relationship between latent concepts and visual content among abundant unimodal data enables GTL to effectively transfer knowledge from unimodal to novel multimodal data, as humans did. Comprehensive experiments demonstrate that the GTL achieves state-of-the-art performance across seven multi-modal datasets across RGB-Sketch, RGB-Infrared, and RGB-Depth.

Chen Li, Eric Peh, Basura Fernando

Recent advances in large vision-language models (VLMs) have shown significant promise for 3D scene understanding. Existing VLM-based approaches typically align 3D scene features with the VLM's embedding space. However, this implicit alignment often yields suboptimal performance due to the scarcity of 3D data and the inherent complexity of spatial relationships in 3D environments. To address these limitations, we propose a novel hierarchical multimodal representation for 3D scene reasoning that explicitly aligns with VLMs at the input space by leveraging both multi-view images and text descriptions. The text descriptions capture spatial relationships by referencing the 3D coordinates of detected objects, while the multi-view images include a top-down perspective and four directional views (forward, left, right, and backward), ensuring comprehensive scene coverage. Additionally, we introduce a hierarchical feature representation that aggregates patch-level image features into view-level and scene-level representations, enabling the model to reason over both local and global scene context. Experimental results on both situated 3D Q&A and general 3D Q&A benchmarks demonstrate the effectiveness of our approach.

Xinyu Zhang, Yuxuan Dong, Lingling Zhang et al.

Despite significant advances in Vision Language Models (VLMs), they remain constrained by the complexity and redundancy of visual input. When images contain large amounts of irrelevant information, VLMs are susceptible to interference, thus generating excessive task-irrelevant reasoning processes or even hallucinations. This limitation stems from their inability to discover and process the required regions during reasoning precisely. To address this limitation, we present the Chain of Foresight-Focus Thought (CoFFT), a novel training-free approach that enhances VLMs' visual reasoning by emulating human visual cognition. Each Foresight-Focus Thought consists of three stages: (1) Diverse Sample Generation: generates diverse reasoning samples to explore potential reasoning paths, where each sample contains several reasoning steps; (2) Dual Foresight Decoding: rigorously evaluates these samples based on both visual focus and reasoning progression, adding the first step of optimal sample to the reasoning process; (3) Visual Focus Adjustment: precisely adjust visual focus toward regions most beneficial for future reasoning, before returning to stage (1) to generate subsequent reasoning samples until reaching the final answer. These stages function iteratively, creating an interdependent cycle where reasoning guides visual focus and visual focus informs subsequent reasoning. Empirical results across multiple benchmarks using Qwen2.5-VL, InternVL-2.5, and Llava-Next demonstrate consistent performance improvements of 3.1-5.8% with controllable increasing computational overhead.

Paritosh Parmar, Eric Peh, Basura Fernando

Existing Causal-Why Video Question Answering (VideoQA) models often struggle with higher-order reasoning, relying on opaque, monolithic pipelines that entangle video understanding, causal inference, and answer generation. These black-box approaches offer limited interpretability and tend to depend on shallow heuristics. We propose a novel, modular framework that explicitly decouples causal reasoning from answer generation, introducing natural language causal chains as interpretable intermediate representations. Inspired by human cognitive models, these structured cause-effect sequences bridge low-level video content with high-level causal reasoning, enabling transparent and logically coherent inference. Our two-stage architecture comprises a Causal Chain Extractor (CCE) that generates causal chains from video-question pairs, and a Causal Chain-Driven Answerer (CCDA) that produces answers grounded in these chains. To address the lack of annotated reasoning traces, we introduce a scalable method for generating high-quality causal chains from existing datasets using large language models. We also propose CauCo, a new evaluation metric for causality-oriented captioning. Experiments on three large-scale benchmarks demonstrate that our approach not only outperforms state-of-the-art models, but also yields substantial gains in explainability, user trust, and generalization -- positioning the CCE as a reusable causal reasoning engine across diverse domains. Project page: https://paritoshparmar.github.io/chainreaction/

Hao Zhang, Chen Li, Basura Fernando

In this early study, we observe an Easy-Options Bias (EOB) issue in some multiple-choice Visual Question Answering (VQA) benchmarks such as MMStar, RealWorldQA, SEED-Bench, Next-QA, STAR benchmark and Video-MME. This bias allows vision-language models (VLMs) to select the correct answer using only the vision (V) and options (O) as inputs, without the need for the question (Q). Through grounding experiments, we attribute the bias to an imbalance in visual relevance: the correct answer typically aligns more closely with the visual contents than the negative options in feature space, creating a shortcut for VLMs to infer the answer via simply vision-option similarity matching. To fix this, we introduce GroundAttack, a toolkit that automatically generates hard negative options as visually plausible as the correct answer. We apply it to the NExT-QA and MMStar datasets, creating new EOB-free annotations. On these EOB-free annotations, current VLMs approach to random accuracies under (V+O) settings, and drop to non-saturated accuracies under (V+Q+O) settings, providing a more realistic evaluation of VLMs' QA ability. Codes and new annotations will be released soon.

Ramanathan Rajendiran, Debaditya Roy, Basura Fernando

Anticipating future events is crucial for various application domains such as healthcare, smart home technology, and surveillance. Narrative event descriptions provide context-rich information, enhancing a system's future planning and decision-making capabilities. We propose a novel task: $\textit{long-term future narration generation}$, which extends beyond traditional action anticipation by generating detailed narrations of future daily activities. We introduce a visual-language model, ViNa, specifically designed to address this challenging task. ViNa integrates long-term videos and corresponding narrations to generate a sequence of future narrations that predict subsequent events and actions over extended time horizons. ViNa extends existing multimodal models that perform only short-term predictions or describe observed videos by generating long-term future narrations for a broader range of daily activities. We also present a novel downstream application that leverages the generated narrations called future video retrieval to help users improve planning for a task by visualizing the future. We evaluate future narration generation on the largest egocentric dataset Ego4D.

Xinyu Zhang, Lingling Zhang, Yanrui Wu et al.

Visual Question Generation (VQG) research focuses predominantly on natural images while neglecting the diagram, which is a critical component in educational materials. To meet the needs of pedagogical assessment, we propose the Diagram-Driven Course Questions Generation (DDCQG) task and construct DiagramQG, a comprehensive dataset with 15,720 diagrams and 25,798 questions across 37 subjects and 371 courses. Our approach employs course and input text constraints to generate course-relevant questions about specific diagram elements. We reveal three challenges of DDCQG: domain-specific knowledge requirements across courses, long-tail distribution in course coverage, and high information density in diagrams. To address these, we propose the Hierarchical Knowledge Integration framework (HKI-DDCQG), which utilizes trainable CLIP for identifying relevant diagram patches, leverages frozen vision-language models for knowledge extraction, and generates questions with trainable T5. Experiments demonstrate that HKI-DDCQG outperforms existing models on DiagramQG while maintaining strong generalizability across natural image datasets, establishing a strong baseline for DDCQG.

Ee Yeo Keat, Zhang Hao, Alexander Matyasko et al.

We introduce VidTFS, a Training-free, open-vocabulary video goal and action inference framework that combines the frozen vision foundational model (VFM) and large language model (LLM) with a novel dynamic Frame Selection module. Our experiments demonstrate that the proposed frame selection module improves the performance of the framework significantly. We validate the performance of the proposed VidTFS on four widely used video datasets, including CrossTask, COIN, UCF101, and ActivityNet, covering goal inference and action recognition tasks under open-vocabulary settings without requiring any training or fine-tuning. The results show that VidTFS outperforms pretrained and instruction-tuned multimodal language models that directly stack LLM and VFM for downstream video inference tasks. Our VidTFS with its adaptability shows the future potential for generalizing to new training-free video inference tasks.

Paritosh Parmar, Eric Peh, Basura Fernando

We introduce the novel concept of visually Connecting Actions and Their Effects (CATE) in video understanding. CATE can have applications in areas like task planning and learning from demonstration. We identify and explore two different aspects of the concept of CATE: Action Selection (AS) and Effect-Affinity Assessment (EAA), where video understanding models connect actions and effects at semantic and fine-grained levels, respectively. We design various baseline models for AS and EAA. Despite the intuitive nature of the task, we observe that models struggle, and humans outperform them by a large margin. Our experiments show that in solving AS and EAA, models learn intuitive properties like object tracking and pose encoding without explicit supervision. We demonstrate that CATE can be an effective self-supervised task for learning video representations from unlabeled videos. The study aims to showcase the fundamental nature and versatility of CATE, with the hope of inspiring advanced formulations and models.

Ramanathan Rajendiran, Debaditya Roy, Basura Fernando

Humans have the natural ability to recognize actions even if the objects involved in the action or the background are changed. Humans can abstract away the action from the appearance of the objects which is referred to as compositionality of actions. We focus on this compositional aspect of action recognition to impart human-like generalization abilities to video action-recognition models. First, we propose an interaction model that captures both fine-grained and long-range interactions between hands and objects. Frame-wise hand-object interactions capture fine-grained movements, while long-range interactions capture broader context and disambiguate actions across time. Second, in order to provide additional contextual cues to differentiate similar actions, we infuse the interaction tokens with global motion information from video tokens. The final global motion refined interaction tokens are used for compositional action recognition. We show the effectiveness of our interaction-centric approach on the compositional Something-Else dataset where we obtain a new state-of-the-art result outperforming recent object-centric methods by a significant margin.

Yunlu Chen, Basura Fernando, Hakan Bilen et al.

In recent years, neural implicit representations have made remarkable progress in modeling of 3D shapes with arbitrary topology. In this work, we address two key limitations of such representations, in failing to capture local 3D geometric fine details, and to learn from and generalize to shapes with unseen 3D transformations. To this end, we introduce a novel family of graph implicit functions with equivariant layers that facilitates modeling fine local details and guaranteed robustness to various groups of geometric transformations, through local $k$-NN graph embeddings with sparse point set observations at multiple resolutions. Our method improves over the existing rotation-equivariant implicit function from 0.69 to 0.89 (IoU) on the ShapeNet reconstruction task. We also show that our equivariant implicit function can be extended to other types of similarity transformations and generalizes to unseen translations and scaling.

Cristian Rodriguez-Opazo, Edison Marrese-Taylor, Basura Fernando et al.

We propose LocFormer, a Transformer-based model for video grounding which operates at a constant memory footprint regardless of the video length, i.e. number of frames. LocFormer is designed for tasks where it is necessary to process the entire long video and at its core lie two main contributions. First, our model incorporates a new sampling technique that splits the input feature sequence into a fixed number of sections and selects a single feature per section using a stochastic approach, which allows us to obtain a feature sample set that is representative of the video content for the task at hand while keeping the memory footprint constant. Second, we propose a modular design that separates functionality, enabling us to learn an inductive bias via supervising the self-attention heads, while also effectively leveraging pre-trained text and video encoders. We test our proposals on relevant benchmark datasets for video grounding, showing that not only LocFormer can achieve excellent results including state-of-the-art performance on YouCookII, but also that our sampling technique is more effective than competing counterparts and that it consistently improves the performance of prior work, by up to 3.13\% in the mean temporal IoU, ultimately leading to a new state-of-the-art performance on Charades-STA.

Arushi Goel, Basura Fernando, Frank Keller et al.

Scene graph generation (SGG) aims to capture a wide variety of interactions between pairs of objects, which is essential for full scene understanding. Existing SGG methods trained on the entire set of relations fail to acquire complex reasoning about visual and textual correlations due to various biases in training data. Learning on trivial relations that indicate generic spatial configuration like 'on' instead of informative relations such as 'parked on' does not enforce this complex reasoning, harming generalization. To address this problem, we propose a novel framework for SGG training that exploits relation labels based on their informativeness. Our model-agnostic training procedure imputes missing informative relations for less informative samples in the training data and trains a SGG model on the imputed labels along with existing annotations. We show that this approach can successfully be used in conjunction with state-of-the-art SGG methods and improves their performance significantly in multiple metrics on the standard Visual Genome benchmark. Furthermore, we obtain considerable improvements for unseen triplets in a more challenging zero-shot setting.

Yan Bin Ng, Basura Fernando

We present a new architecture for human action forecasting from videos. A temporal recurrent encoder captures temporal information of input videos while a self-attention model is used to attend on relevant feature dimensions of the input space. To handle temporal variations in observed video data, a feature masking techniques is employed. We classify observed actions accurately using an auxiliary classifier which helps to understand what has happened so far. Then the decoder generates actions for the future based on the output of the recurrent encoder and the self-attention model. Experimentally, we validate each component of our architecture where we see that the impact of self-attention to identify relevant feature dimensions, temporal masking, and observed auxiliary classifier. We evaluate our method on two standard action forecasting benchmarks and obtain state-of-the-art results.

Basura Fernando, Samitha Herath

We propose a framework for early action recognition and anticipation by correlating past features with the future using three novel similarity measures called Jaccard vector similarity, Jaccard cross-correlation and Jaccard Frobenius inner product over covariances. Using these combinations of novel losses and using our framework, we obtain state-of-the-art results for early action recognition in UCF101 and JHMDB datasets by obtaining 91.7 % and 83.5 % accuracy respectively for an observation percentage of 20. Similarly, we obtain state-of-the-art results for Epic-Kitchen55 and Breakfast datasets for action anticipation by obtaining 20.35 and 41.8 top-1 accuracy respectively.

Haziq Razali, Basura Fernando

The use of latent variable models has shown to be a powerful tool for modeling probability distributions over sequences. In this paper, we introduce a new variational model that extends the recurrent network in two ways for the task of video frame prediction. First, we introduce 3D convolutions inside all modules including the recurrent model for future frame prediction, inputting and outputting a sequence of video frames at each timestep. This enables us to better exploit spatiotemporal information inside the variational recurrent model, allowing us to generate high-quality predictions. Second, we enhance the latent loss of the variational model by introducing a maximum likelihood estimate in addition to the KL divergence that is commonly used in variational models. This simple extension acts as a stronger regularizer in the variational autoencoder loss function and lets us obtain better results and generalizability. Experiments show that our model outperforms existing video prediction methods on several benchmarks while requiring fewer parameters.

Vinoj Jayasundara, Debaditya Roy, Basura Fernando

Capsule networks (CapsNets) have recently shown promise to excel in most computer vision tasks, especially pertaining to scene understanding. In this paper, we explore CapsNet's capabilities in optical flow estimation, a task at which convolutional neural networks (CNNs) have already outperformed other approaches. We propose a CapsNet-based architecture, termed FlowCaps, which attempts to a) achieve better correspondence matching via finer-grained, motion-specific, and more-interpretable encoding crucial for optical flow estimation, b) perform better-generalizable optical flow estimation, c) utilize lesser ground truth data, and d) significantly reduce the computational complexity in achieving good performance, in comparison to its CNN-counterparts.

Cristian Rodriguez-Opazo, Edison Marrese-Taylor, Basura Fernando et al.

This paper studies the task of temporal moment localization in a long untrimmed video using natural language query. Given a query sentence, the goal is to determine the start and end of the relevant segment within the video. Our key innovation is to learn a video feature embedding through a language-conditioned message-passing algorithm suitable for temporal moment localization which captures the relationships between humans, objects and activities in the video. These relationships are obtained by a spatial sub-graph that contextualizes the scene representation using detected objects and human features conditioned in the language query. Moreover, a temporal sub-graph captures the activities within the video through time. Our method is evaluated on three standard benchmark datasets, and we also introduce YouCookII as a new benchmark for this task. Experiments show our method outperforms state-of-the-art methods on these datasets, confirming the effectiveness of our approach.

Rodrigo Santa Cruz, Anoop Cherian, Basura Fernando et al.

This paper presents a framework to recognize temporal compositions of atomic actions in videos. Specifically, we propose to express temporal compositions of actions as semantic regular expressions and derive an inference framework using probabilistic automata to recognize complex actions as satisfying these expressions on the input video features. Our approach is different from existing works that either predict long-range complex activities as unordered sets of atomic actions, or retrieve videos using natural language sentences. Instead, the proposed approach allows recognizing complex fine-grained activities using only pretrained action classifiers, without requiring any additional data, annotations or neural network training. To evaluate the potential of our approach, we provide experiments on synthetic datasets and challenging real action recognition datasets, such as MultiTHUMOS and Charades. We conclude that the proposed approach can extend state-of-the-art primitive action classifiers to vastly more complex activities without large performance degradation.

Yan Bin Ng, Basura Fernando

Future human action forecasting from partial observations of activities is an important problem in many practical applications such as assistive robotics, video surveillance and security. We present a method to forecast actions for the unseen future of the video using a neural machine translation technique that uses encoder-decoder architecture. The input to this model is the observed RGB video, and the objective is to forecast the correct future symbolic action sequence. Unlike prior methods that make action predictions for some unseen percentage of video one for each frame, we predict the complete action sequence that is required to accomplish the activity. We coin this task action sequence forecasting. To cater for two types of uncertainty in the future predictions, we propose a novel loss function. We show a combination of optimal transport and future uncertainty losses help to improve results. We extend our action sequence forecasting model to perform weakly supervised action forecasting on two challenging datasets, the Breakfast and the 50Salads. Specifically, we propose a model to predict actions of future unseen frames without using frame level annotations during training. Using Fisher vector features, our supervised model outperforms the state-of-the-art action forecasting model by 0.83% and 7.09% on the Breakfast and the 50Salads datasets respectively. Our weakly supervised model is only 0.6% behind the most recent state-of-the-art supervised model and obtains comparable results to other published fully supervised methods, and sometimes even outperforms them on the Breakfast dataset. Most interestingly, our weakly supervised model outperforms prior models by 1.04% leveraging on proposed weakly supervised architecture, and effective use of attention mechanism and loss functions.

Arushi Goel, Basura Fernando, Thanh-Son Nguyen et al.

Automatically generating natural language descriptions from an image is a challenging problem in artificial intelligence that requires a good understanding of the visual and textual signals and the correlations between them. The state-of-the-art methods in image captioning struggles to approach human level performance, especially when data is limited. In this paper, we propose to improve the performance of the state-of-the-art image captioning models by incorporating two sources of prior knowledge: (i) a conditional latent topic attention, that uses a set of latent variables (topics) as an anchor to generate highly probable words and, (ii) a regularization technique that exploits the inductive biases in syntactic and semantic structure of captions and improves the generalization of image captioning models. Our experiments validate that our method produces more human interpretable captions and also leads to significant improvements on the MSCOCO dataset in both the full and low data regimes.

Yuge Shi, Basura Fernando, Richard Hartley

We introduce a novel Recurrent Neural Network-based algorithm for future video feature generation and action anticipation called feature mapping RNN. Our novel RNN architecture builds upon three effective principles of machine learning, namely parameter sharing, Radial Basis Function kernels and adversarial training. Using only some of the earliest frames of a video, the feature mapping RNN is able to generate future features with a fraction of the parameters needed in traditional RNN. By feeding these future features into a simple multi-layer perceptron facilitated with an RBF kernel layer, we are able to accurately predict the action in the video. In our experiments, we obtain 18% improvement on JHMDB-21 dataset, 6% on UCF101-24 and 13% improvement on UT-Interaction datasets over prior state-of-the-art for action anticipation.

Yan Bin Ng, Basura Fernando

This paper classifies human action sequences from videos using a machine translation model. In contrast to classical human action classification which outputs a set of actions, our method output a sequence of action in the chronological order of the actions performed by the human. Therefore our method is evaluated using sequential performance measures such as Bilingual Evaluation Understudy (BLEU) scores. Action sequence classification has many applications such as learning from demonstration, action segmentation, detection, localization and video captioning. Furthermore, we use our model that is trained to output action sequences to solve downstream tasks; such as video captioning and action localization. We obtain state of the art results for video captioning in challenging Charades dataset obtaining BLEU-4 score of 34.8 and METEOR score of 33.6 outperforming previous state-of-the-art of 18.8 and 19.5 respectively. Similarly, on ActivityNet captioning, we obtain excellent results in-terms of ROUGE (20.24) and CIDER (37.58) scores. For action localization, without using any explicit start/end action annotations, our method obtains localization performance of 22.2 mAP outperforming prior fully supervised methods.

Basura Fernando, Cheston Tan Yin Chet, Hakan Bilen

Detecting temporal extents of human actions in videos is a challenging computer vision problem that requires detailed manual supervision including frame-level labels. This expensive annotation process limits deploying action detectors to a limited number of categories. We propose a novel method, called WSGN, that learns to detect actions from \emph{weak supervision}, using only video-level labels. WSGN learns to exploit both video-specific and dataset-wide statistics to predict relevance of each frame to an action category. This strategy leads to significant gains in action detection for two standard benchmarks THUMOS14 and Charades. Our method obtains excellent results compared to state-of-the-art methods that uses similar features and loss functions on THUMOS14 dataset. Similarly, our weakly supervised method is only 0.3% mAP behind a state-of-the-art supervised method on challenging Charades dataset for action localization.

Mohammad Sadegh Aliakbarian, Fatemeh Sadat Saleh, Mathieu Salzmann et al.

Action anticipation is critical in scenarios where one needs to react before the action is finalized. This is, for instance, the case in automated driving, where a car needs to, e.g., avoid hitting pedestrians and respect traffic lights. While solutions have been proposed to tackle subsets of the driving anticipation tasks, by making use of diverse, task-specific sensors, there is no single dataset or framework that addresses them all in a consistent manner. In this paper, we therefore introduce a new, large-scale dataset, called VIENA2, covering 5 generic driving scenarios, with a total of 25 distinct action classes. It contains more than 15K full HD, 5s long videos acquired in various driving conditions, weathers, daytimes and environments, complemented with a common and realistic set of sensor measurements. This amounts to more than 2.25M frames, each annotated with an action label, corresponding to 600 samples per action class. We discuss our data acquisition strategy and the statistics of our dataset, and benchmark state-of-the-art action anticipation techniques, including a new multi-modal LSTM architecture with an effective loss function for action anticipation in driving scenarios.

Cristian Rodriguez, Basura Fernando, Hongdong Li

Human action-anticipation methods predict what is the future action by observing only a few portion of an action in progress. This is critical for applications where computers have to react to human actions as early as possible such as autonomous driving, human-robotic interaction, assistive robotics among others. In this paper, we present a method for human action anticipation by predicting the most plausible future human motion. We represent human motion using Dynamic Images and make use of tailored loss functions to encourage a generative model to produce accurate future motion prediction. Our method outperforms the currently best performing action-anticipation methods by 4% on JHMDB-21, 5.2% on UT-Interaction and 5.1% on UCF 101-24 benchmarks.

Rodrigo Santa Cruz, Basura Fernando, Anoop Cherian et al.

The world is fundamentally compositional, so it is natural to think of visual recognition as the recognition of basic visually primitives that are composed according to well-defined rules. This strategy allows us to recognize unseen complex concepts from simple visual primitives. However, the current trend in visual recognition follows a data greedy approach where huge amounts of data are required to learn models for any desired visual concept. In this paper, we build on the compositionality principle and develop an "algebra" to compose classifiers for complex visual concepts. To this end, we learn neural network modules to perform boolean algebra operations on simple visual classifiers. Since these modules form a complete functional set, a classifier for any complex visual concept defined as a boolean expression of primitives can be obtained by recursively applying the learned modules, even if we do not have a single training sample. As our experiments show, using such a framework, we can compose classifiers for complex visual concepts outperforming standard baselines on two well-known visual recognition benchmarks. Finally, we present a qualitative analysis of our method and its properties.