Adaptive Wing Loss for Robust Face Alignment via Heatmap Regression

By Xinyao Wang, Liefeng Bo, Fuxin Li. IEEE International Conference on Computer Vision, 2019

301 citations recorded by the source corpus.

Heatmap regression with a deep network has become one of the mainstream approaches to localize facial landmarks. However, the loss function for heatmap regression is rarely studied. In this paper, we analyze the ideal loss function properties for heatmap regression in face alignment problems. Then we propose a novel loss function, named Adaptive Wing loss, that is able to adapt its shape to different types of ground truth heatmap pixels. This adaptability penalizes loss more on foreground pixels while less on background pixels. To address the imbalance between foreground and background pixels, we also propose Weighted Loss Map, which assigns high weights on foreground and difficult background pixels to help training process focus more on pixels that are crucial to landmark localization. To further improve face alignment accuracy, we introduce boundary prediction and CoordConv with boundary coordinates. Extensive experiments on different benchmarks, including COFW, 300W and WFLW, show our approach outperforms the state-of-the-art by a significant margin on various evaluation metrics. Besides, the Adaptive Wing loss also helps other heatmap regression tasks.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/27aae8c3d70b0fec180d5be2c66269adb140c73a

How researchers see this paper

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

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

StarGAN v2: Diverse Image Synthesis for Multiple Domains (2019, methodology): “We also remove all shortcuts in the upsampling residual blocks and add skip connections with the adaptive wing based heatmap [46].”
3FabRec: Fast Few-Shot Face Alignment by Reconstruction (2019, methodology): “The currently best-performing method on many benchmarks uses a heatmap-based framework together with optimization of the loss function to foreground versus background pixels [47].”
General Facial Representation Learning in a Visual-Linguistic Manner (2021, methodology): “Instead of using those complex loss functions designed by [30, 42, 97], we simply train the head with a soft-label cross-entropy loss.”
Autonomous Robotic Laparoscopic Surgery for Intestinal Anastomosis (2022, methodology): “The second U-Net took both the intestine image and the segmentation result from the first U-Net as the input and it output a landmark heatmap, because the landmark heatmap regression has demonstrated better performances than regressing the landmark coordinates directly (42).”