サインイン
インサイト - Affective Computing
VLLMs Leverage Common Sense Reasoning to Enhance Emotion Understanding in Context
Leveraging the common sense reasoning capabilities of Vision-and-Large-Language Models (VLLMs), this work proposes a novel two-stage approach to enhance emotion classification in visual context without introducing complex training pipelines.
AU-TGN: Advancing Action Unit Detection with Temporal Convolution and GPT-2 in Wild Audiovisual Contexts
音声と視覚データのシナジーを活用し、AU検出の精度向上を図る新しい手法を提案。
Analyzing Prompt Sensitivity of ChatGPT in Affective Computing
Prompt sensitivity analysis in ChatGPT for affective computing tasks reveals the impact of different prompts and generation parameters on model performance.
Audio-Visual Compound Expression Recognition Method for Affective Computing
Proposing a novel audio-visual method for compound expression recognition based on emotion probability fusion and rule-based decision-making.
Emotion-Aware Multimodal Fusion for Meme Emotion Detection: A Novel Approach
Introducing ALFRED, a novel multimodal neural framework for meme emotion detection, outperforming existing baselines by 4.94% F1.
© 2024 by Linnk AI