Facial Expression Recognition with Inconsistently Annotated Datasets

By Jiabei Zeng, Shiguang Shan, Xilin Chen. European Conference on Computer Vision, 2018

315 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Facial Expression Recognition With Visual Transformers and Attentional Selective Fusion (2021, methodology): “IPA2LT [60], gACNN [9] and SPWFA-SE [56] are trained for seven classes on AffectNet without the contempt category.”
Region Attention Networks for Pose and Occlusion Robust Facial Expression Recognition (2019, methodology): “It is worth noting that [44] only achieves 47% with upsampling and [58] uses one more large-scale FER dataset and 80 layers ResNet for training with elaborated loss weights on them.”
Suppressing Uncertainties for Large-Scale Facial Expression Recognition (2020, background): “IPA2LT [43] introduces the latent ground-truth idea for training with inconsistent annotations across different FER datasets. gaCNN [24] leverages a patch-based attention network and a global network.”
IExpressNet: Facial Expression Recognition with Incremental Classes (2020, methodology): “Comparing these two models, CES model is more consistent with daily usage [41, 51, 73, 82], which is often used in real applications.”
Robust Lightweight Facial Expression Recognition Network with Label Distribution Training (2021, methodology): “For the methods of IPA2LT (Zeng, Shan, and Chen 2018) and gACNN (Li et al. 2019c), only the lower limit values are provided (the symbol > is used).”
Deep Facial Expression Recognition: A Survey (2018, background): “cannot keep improving when enlarging the training data by directly merging multiple datasets [129].”