AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Facial Affective Behavior Analysis with Instruction Tuning

Yifan Li, Anh Dao, Wentao Bao, Zhen Tan, Tianlong Chen, Huan Liu, Yu Kong

ECCV 2024 · Vision-language and multimodal › Multimodal LLMs · approach: Benchmarks and analysis of MLLMs · 20 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add facial landmark prior token alongside visual tokens (ablation)EmoLA · FABA-Bench emotionrecognition accuracy S_re: 62.5 -> 64.5Improved
add facial landmark prior token alongside visual tokens (ablation)EmoLA · FABA-Bench AUrecognition F1 S_re: 55.3 -> 56.3Improved
add visual tokens to the prior-only input (ablation)EmoLA · FABA-Bench emotionrecognition accuracy S_re: 41.2 -> 62.5Improved
tune LoRA with visual projector in addition to prior projector (ablation)EmoLA · FABA-Bench emotionrecognition accuracy S_re: 44.9 -> 64.5Improved
also tune prior projector on top of LoRA and visual projector (ablation)EmoLA · FABA-Bench AUrecognition F1 S_re: 55.6 -> 56.3Improved
multi-task training on emotion and AU instead of single task (ablation)EmoLA · FABA-Bench AUREGE: 91.5 -> 88.6Hurt
multi-task training on emotion and AU instead of single task (ablation)EmoLA · FABA-Bench emotionREGE: 96.2 -> 96.3No clear effect
place prior token after visual tokens rather than before (ablation)EmoLA · FABA-Bench AUREGE: 89.7 -> 91.5Improved
use EmoLA (LoRA, facial prior) instead of fully fine-tuned LLaVA-1.5FABA-Bench emotion · FABA-Bench emotionREGE: 93.9 -> 96.2Improved
use EmoLA instead of ReCoT on BP4DBP4D · BP4Daverage F1 (%): 64.8 -> 64.2Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy