Research map / Vision-language and multimodal / Human and embodied modeling
EmoSet: A Large-scale Visual Emotion Dataset with Rich Attributes
Jingyuan Yang, Qirui Huang, Tingting Ding, Dani Lischinski, Danny Cohen-Or, Hui Huang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add attribute module (brightness, scene type, facial expression) to backbone (ablation) | ResNet-50 trained from scratch · EmoSet-8 | top-1 accuracy (%): 51.48 -> 58.62 | Improved |
| add attribute module (ablation) | ImageNet-pretrained backbones · EmoSet-8 | top-1 accuracy (%): average 71.61 -> 74.10 | Improved |
| add attribute module (ablation) | backbones trained from scratch · EmoSet-8 | top-1 accuracy (%): average 49.88 -> 57.93 | Improved |
| ImageNet pretraining of backbone (ablation) | ResNet-50 · EmoSet-8 | top-1 accuracy (%): 51.48 -> 74.04 | Improved |
| train on EmoSet instead of FI when testing on FI | ResNet-50 · FI test (pretrained) | top-1 accuracy (%): 67.53 -> 55.67 | Hurt |
| train on FI instead of EmoSet when testing on EmoSet | ResNet-50 · EmoSet test (pretrained) | top-1 accuracy (%): 74.04 -> 53.95 | Hurt |
| train on EmoSet instead of FI, test on ArtPhoto | ResNet-50 · ArtPhoto test (pretrained) | top-1 accuracy (%): 29.28 -> 32.26 | Improved |
| brighter and more colorful images | emotion polarity · EmoSet-118K | share of positive emotion labels | Improved |
Closest papers by meaning
- EmoNet-Face: An Expert-Annotated Benchmark for Synthetic Emotion Recognition (NeurIPS 2025 · similarity 0.90)
- EEmo-Logic: A Unified Dataset and Multi-Stage Framework for Comprehensive Image-Evoked Emotion Assessment (ICML 2026 · similarity 0.89)
- Customizing Visual Emotion Evaluation for MLLMs: An Open-vocabulary, Multifaceted, and Scalable Approach (ICLR 2026 · similarity 0.89)
- Bridge Then Begin Anew: Generating Target-Relevant Intermediate Model for Source-Free Visual Emotion Adaptation (AAAI 2025 · similarity 0.89)
- $E^3$: Exploring Embodied Emotion Through A Large-Scale Egocentric Video Dataset (NeurIPS 2024 · similarity 0.89)
- EmoEdit: Evoking Emotions through Image Manipulation (CVPR 2025 · similarity 0.88)
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.
