MétaCan
Menu
Back to cohort
Record W4301018292 · doi:10.48550/arxiv.1510.02969

Do Deep Neural Networks Learn Facial Action Units When Doing Expression\n Recognition?

2015· preprint· en· W4301018292 on OpenAlexaboutno aff
Pooya Khorrami, Tom Le Paine, Thomas S. Huang

Bibliographic record

VenuearXiv (Cornell University) · 2015
Typepreprint
Languageen
FieldComputer Science
TopicFace recognition and analysis
Canadian institutionsnot available
Fundersnot available
KeywordsConvolutional neural networkComputer scienceClassifier (UML)Artificial intelligencePattern recognition (psychology)Facial expressionFacial expression recognitionFacial recognition systemFace (sociological concept)Action recognitionExpression (computer science)Speech recognition

Abstract

fetched live from OpenAlex

Despite being the appearance-based classifier of choice in recent years,\nrelatively few works have examined how much convolutional neural networks\n(CNNs) can improve performance on accepted expression recognition benchmarks\nand, more importantly, examine what it is they actually learn. In this work,\nnot only do we show that CNNs can achieve strong performance, but we also\nintroduce an approach to decipher which portions of the face influence the\nCNN's predictions. First, we train a zero-bias CNN on facial expression data\nand achieve, to our knowledge, state-of-the-art performance on two expression\nrecognition benchmarks: the extended Cohn-Kanade (CK+) dataset and the Toronto\nFace Dataset (TFD). We then qualitatively analyze the network by visualizing\nthe spatial patterns that maximally excite different neurons in the\nconvolutional layers and show how they resemble Facial Action Units (FAUs).\nFinally, we use the FAU labels provided in the CK+ dataset to verify that the\nFAUs observed in our filter visualizations indeed align with the subject's\nfacial movements.\n

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMeta-epidemiology (narrow)
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.909
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.001
Science and technology studies0.0000.000
Scholarly communication0.0000.001
Open science0.0010.001
Research integrity0.0000.001
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.142
GPT teacher head0.218
Teacher spread0.076 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

Study designSimulation or modeling
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2015
Admission routes1
Has abstractyes

Explore more

Same venuearXiv (Cornell University)Same topicFace recognition and analysisFrench-language works237,207