Multi-View Exclusive Unsupervised Dimension Reduction for Video-Based Facial Expression Recognition

Authors: Liping Xie, Dacheng Tao, Haikun Wei

IJCAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on two challenging video-based FER datasets demonstrate the effectiveness of the proposed method.
Researcher Affiliation Academia Key Laboratory of Measurement and Control of CSE, Ministry of Education, School of Automation, Southeast University, Nanjing 210096, China QCIS and FEIT, University of Technology Sydney, Australia
Pseudocode Yes Algorithm 1 The efficient iterative algorithm for solving Eq.(6)
Open Source Code No No explicit statement about providing open-source code for the described methodology was found.
Open Datasets Yes The first dataset is the facial expression (FE) dataset proposed in [Doll ar et al., 2005], which we call FE05 in this paper. The second FE dataset is the Oulu-CASIA VIS (CASIA for short) database [Li et al., 2013].
Dataset Splits Yes The five-fold cross-validation strategy is adopted for both datasets for tuning the parameters. We randomly separate these subjects into two groups: 70 subjects with six expressions (420 samples) for training, 10 subjects (60 samples) for testing.
Hardware Specification No No specific hardware details (like GPU/CPU models, memory, or cloud instances) used for running the experiments were provided in the paper.
Software Dependencies No The paper mentions classifiers like SVM and kNN, but does not provide specific version numbers for any software, libraries, or programming languages used.
Experiment Setup Yes The five-fold cross-validation strategy is adopted for both datasets for tuning the parameters. The candidate set for both trade-off parameters γ and is {10i|i = 5, 4, ..., 3, 4}. Accuracy is averaged over five runs for each dimension r in {1, 2, 5, 8, 10, 15, 20, 30, 50, 80, 100, 150, 200, 300}.