Attributes-aware Visual Emotion Representation Learning
Anonymous Authors, 1 st Rahul Singh Maharjan, 2 nd Marta Romeo, 3 rd Angelo Cangelosi · 2025
Read Attributes-aware Visual Emotion Representation Learning by Anonymous Authors, 1 st Rahul Singh Maharjan, 2 nd Marta Romeo, 3 rd Angelo Cangelosi (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-06578.txt.
Description
Rahul Singh Maharjan, Marta Romeo, and Angelo Cangelosi build a network in 2025 that reads feeling from images. Pictures stir emotions through many cues at once, and the gap between raw pixels and felt emotion is wide. Most past models read the whole image and skipped the cues that drive the feeling.
Their network, A4Net, tracks four cues: brightness, color richness, scene context, and facial expression. It learns the cues and the emotion call together, each helping the other. Tests across several emotion data sets matched top methods. Heat maps from the network show which regions drove its calls, aiding trust in the results.
Catalog record
| CREATOR | Anonymous Authors, 1 st Rahul Singh Maharjan, 2 nd Marta Romeo, 3 rd Angelo Cangelosi |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 37 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-06578