Towards Regularized Mixture of Predictions for Class-Imbalanced Semi-Supervised Facial Expression Recognition
Authors: Hangyu Li, Yixin Zhang, Jiangchao Yao, Nannan Wang, Bo Han
IJCAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on four facial expression datasets demonstrate the effectiveness of the proposed method across various imbalanced conditions. The source code is made publicly available at https://github. com/hangyu94/Re Mo P. 5 Experiments |
| Researcher Affiliation | Academia | 1TMLR Group, Department of Computer Science, Hong Kong Baptist University, Hong Kong, China 2State Key Laboratory of Integrated Services Networks, Xidian University, Xi an, China 3Cooperative Medianet Innovation Center, Shanghai Jiao Tong University, Shanghai, China EMAIL, EMAIL, EMAIL, EMAIL |
| Pseudocode | Yes | Algorithm 1 Re Mo P s main learning algorithm. |
| Open Source Code | Yes | The source code is made publicly available at https://github. com/hangyu94/Re Mo P. |
| Open Datasets | Yes | We conduct experiments on four public facial expression datasets, including RAF-DB [Li and Deng, 2019], FERPlus [Barsoum et al., 2016], CK+ [Lucey et al., 2010], and Affect Net [Mollahosseini et al., 2017]. |
| Dataset Splits | Yes | In our experiments, we use the single-label subset with 12,271 training images and 3,068 testing images, including seven basic facial expression categories (i.e., surprise, fear, disgust, happiness, sadness, anger, and neutral). FERPlus provides the new eight-class labels (i.e., seven basic categories and contempt) created by 10 crowd-sourced annotators. It consists of 28,709 training images, 3,589 validation images, and 3,589 testing images. |
| Hardware Specification | Yes | We implement all experiments using the Py Torch toolbox with one NVIDIA A100 GPU. |
| Software Dependencies | No | We implement all experiments using the Py Torch toolbox with one NVIDIA A100 GPU. For the basic encoder, we use the Res Net-18 [He et al., 2016] pre-trained on the MS-Celeb-1M face recognition dataset [Guo et al., 2016] for learning facial expression features. Besides, we use MTCNN [Zhang et al., 2016] to align and resize facial images to 224 224 pixels. ... By default, we use the Adam optimizer. |
| Experiment Setup | Yes | By default, we use the Adam optimizer. For RAF-DB and CK+, we train the model with a learning rate of 1e 4, training epoch 100, and batch size 16. For FERPlus, due to a large number of training samples, we train the model with a learning rate of 1e 4, training epoch 80, and batch size 32. The number of training iterations tmax is set to 1,000 in all experiments. The relative ratio µ is set to 1 except for Affect Net as 5. The hyper-parameter β in Eq. (6) is set to 0.5. Following Fix Match [Sohn et al., 2020], the default threshold τ is 0.95. In Eq. (9), the hyper-parameters λ1 and λ2 are set to 1e 4 and 0.1, respectively. |