NeuroGaze-Distill: Brain-informed Distillation and Depression-Inspired Geometric Priors for Robust Facial Emotion Recognition

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Li, Zilin, Xu, Weiwei, Zhao, Xuanqi, Zhu, Yiran
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911284210958336
author Li, Zilin
Xu, Weiwei
Zhao, Xuanqi
Zhu, Yiran
author_facet Li, Zilin
Xu, Weiwei
Zhao, Xuanqi
Zhu, Yiran
contents Facial emotion recognition (FER) models trained only on pixels often fail to generalize across datasets because facial appearance is an indirect and biased proxy for underlying affect. We present NeuroGaze-Distill, a cross-modal distillation framework that transfers brain-informed priors into an image-only FER student via static Valence/Arousal (V/A) prototypes and a depression-inspired geometric prior (D-Geo). A teacher trained on EEG topographic maps from DREAMER (with MAHNOB-HCI as unlabeled support) produces a consolidated 5x5 V/A prototype grid that is frozen and reused; no EEG-face pairing and no non-visual signals at deployment are required. The student (ResNet-18/50) is trained on FERPlus with conventional CE/KD and two lightweight regularizers: (i) Proto-KD (cosine) aligns student features to the static prototypes; (ii) D-Geo softly shapes the embedding geometry in line with affective findings often reported in depression research (e.g., anhedonia-like contraction in high-valence regions). We evaluate both within-domain (FERPlus validation) and cross-dataset protocols (AffectNet-mini; optional CK+), reporting standard 8-way scores alongside present-only Macro-F1 and balanced accuracy to fairly handle label-set mismatch. Ablations attribute consistent gains to prototypes and D-Geo, and favor 5x5 over denser grids for stability. The method is simple, deployable, and improves robustness without architectural complexity.
format Preprint
id arxiv_https___arxiv_org_abs_2509_11916
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle NeuroGaze-Distill: Brain-informed Distillation and Depression-Inspired Geometric Priors for Robust Facial Emotion Recognition
Li, Zilin
Xu, Weiwei
Zhao, Xuanqi
Zhu, Yiran
Computer Vision and Pattern Recognition
I.2.10; I.4.8; I.5.4
Facial emotion recognition (FER) models trained only on pixels often fail to generalize across datasets because facial appearance is an indirect and biased proxy for underlying affect. We present NeuroGaze-Distill, a cross-modal distillation framework that transfers brain-informed priors into an image-only FER student via static Valence/Arousal (V/A) prototypes and a depression-inspired geometric prior (D-Geo). A teacher trained on EEG topographic maps from DREAMER (with MAHNOB-HCI as unlabeled support) produces a consolidated 5x5 V/A prototype grid that is frozen and reused; no EEG-face pairing and no non-visual signals at deployment are required. The student (ResNet-18/50) is trained on FERPlus with conventional CE/KD and two lightweight regularizers: (i) Proto-KD (cosine) aligns student features to the static prototypes; (ii) D-Geo softly shapes the embedding geometry in line with affective findings often reported in depression research (e.g., anhedonia-like contraction in high-valence regions). We evaluate both within-domain (FERPlus validation) and cross-dataset protocols (AffectNet-mini; optional CK+), reporting standard 8-way scores alongside present-only Macro-F1 and balanced accuracy to fairly handle label-set mismatch. Ablations attribute consistent gains to prototypes and D-Geo, and favor 5x5 over denser grids for stability. The method is simple, deployable, and improves robustness without architectural complexity.
title NeuroGaze-Distill: Brain-informed Distillation and Depression-Inspired Geometric Priors for Robust Facial Emotion Recognition
topic Computer Vision and Pattern Recognition
I.2.10; I.4.8; I.5.4
url https://arxiv.org/abs/2509.11916