Benchmarking Dynamic Affective Reasoning: A Viewer-Centric Video Emotion Dataset

• URL: http://arxiv.org/abs/2607.10238v1
• Date: Sat, 11 Jul 2026 09:58:29 GMT
• Title: Benchmarking Dynamic Affective Reasoning: A Viewer-Centric Video Emotion Dataset
• Abstract summary: We introduce Dynamic Affective Reasoning, the first large-scale benchmark for viewer-centric affect transitions and causal reasoning over video events.<n>Dar contains 15,087 videos and 36,908 event-aligned affective segments annotated with 27 emotion categories.<n>Dar-R1 sets a new state-of-the-art for dynamic affective reasoning, in terms of both emotional localization and affective reasoning.
• Score: 51.92925649424503
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Video emotion analysis is typically framed as a static classification problem, treating each clip as an independent labeled unit. However, such a formulation overlooks a key psychological fact: emotions change as a result of cumulative reactions to consecutive causal events. To bridge this gap, we introduce Dynamic Affective Reasoning, the first large-scale benchmark for viewer-centric affect transitions and causal reasoning over consecutive video events. DAR contains 15,087 videos and 36,908 event-aligned affective segments annotated with 27 emotion categories. Unlike existing video-based emotion datasets, DAR presents a new viewer-centric perspective on fine-grained emotional expressions and transitions, and provides dense, temporally grounded, and causally explicit reasoning chains. Based on DAR, we formally define three challenging tasks: affective segmentation, fine-grained emotion classification, and affective reasoning. Complementing this benchmark, we propose DAR-R1, a two-stage framework that combines supervised fine-tuning with Group Relative Policy Optimization. Experiments across 10+ MLLMs show that DAR-R1 sets a new state-of-the-art for dynamic affective reasoning, in terms of both emotional localization and affective reasoning. Project page: https://github.com/Zhang-Zhiyan/DAR.

Related papers

• Adaptive Emotional Video Captioning via Affective Heterogeneous Graph Reasoning and Multi-task Joint Learning [26.623474533402575]
  We propose SAGML, an adaptive EVC framework via affective heterogeneous graph and multi-task language modeling.<n> SAGML constructs a soft affective heterogeneous graph containing catalog-level emotion nodes and lexical-level emotion word nodes.<n>It provides an error-resilient and multi-emotion-aware baseline for EVC.
  arXiv  Detail & Related papers  (2026-07-31T05:49:33Z)
• Towards Accurate Emotion-Attributed Video Captioning via Fine-grained Emotion-Cause Pair Extraction [47.762139272076126]
  Emotional Video Captioning (EVC) is a challenging task that aims to generate factually accurate and emotionally rich descriptions for videos.<n>Existing EVC methods leverage holistic visual features to mine global emotional cues, and then aggregate multimodal features to guide the emotional caption generation.<n>We propose a fine-grained emotion-cause pair extraction framework for emotion-attributed video captioning.
  arXiv  Detail & Related papers  (2026-06-07T10:43:41Z)
• FACE-net: Factual Calibration and Emotion Augmentation for Retrieval-enhanced Emotional Video Captioning [81.33341786837974]
  Emotional Video Captioning (EVC) is an emerging task, which aims to describe factual content with the intrinsic emotions expressed in videos.<n>We propose a retrieval-enhanced framework with FActual and Emotion augmentation (FACE-net)<n>FACE-net collaboratively mines factual-emotional semantics and provides adaptive and accurate guidance for generation.
  arXiv  Detail & Related papers  (2026-03-18T07:53:15Z)
• VidEmo: Affective-Tree Reasoning for Emotion-Centric Video Foundation Models [46.591026037722436]
  We propose a novel affective cues-guided reasoning framework that unifies fundamental attribute perception, expression analysis, and high-level emotional understanding.<n>At the core of our approach is a family of video emotion foundation models (VidEmo), specifically designed for emotion reasoning and instruction-following.<n>We establish a foundational data infrastructure and introduce a emotion-centric fine-grained dataset consisting of 2.1M diverse instruction-based samples.
  arXiv  Detail & Related papers  (2025-11-04T16:31:09Z)
• Disentangle Identity, Cooperate Emotion: Correlation-Aware Emotional Talking Portrait Generation [63.94836524433559]
  DICE-Talk is a framework for disentangling identity with emotion and cooperating emotions with similar characteristics.<n>We develop a disentangled emotion embedder that jointly models audio-visual emotional cues through cross-modal attention.<n>Second, we introduce a correlation-enhanced emotion conditioning module with learnable Emotion Banks.<n>Third, we design an emotion discrimination objective that enforces affective consistency during the diffusion process.
  arXiv  Detail & Related papers  (2025-04-25T05:28:21Z)
• Dual-path Collaborative Generation Network for Emotional Video Captioning [33.230028098522254]
  Emotional Video Captioning is an emerging task that aims to describe factual content with the intrinsic emotions expressed in videos. Existing emotional video captioning methods perceive global visual emotional cues at first, and then combine them with the video features to guide the emotional caption generation. We propose a dual-path collaborative generation network, which dynamically perceives visual emotional cues evolutions while generating emotional captions.
  arXiv  Detail & Related papers  (2024-08-06T07:30:53Z)
• A Circular-Structured Representation for Visual Emotion Distribution Learning [82.89776298753661]
  We propose a well-grounded circular-structured representation to utilize the prior knowledge for visual emotion distribution learning. To be specific, we first construct an Emotion Circle to unify any emotional state within it. On the proposed Emotion Circle, each emotion distribution is represented with an emotion vector, which is defined with three attributes.
  arXiv  Detail & Related papers  (2021-06-23T14:53:27Z)
• Affect2MM: Affective Analysis of Multimedia Content Using Emotion Causality [84.69595956853908]
  We present Affect2MM, a learning method for time-series emotion prediction for multimedia content. Our goal is to automatically capture the varying emotions depicted by characters in real-life human-centric situations and behaviors.
  arXiv  Detail & Related papers  (2021-03-11T09:07:25Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.