FSA-Net: Learning Fine-Grained Structure Aggregation for Head Pose Estimation From a Single Image

By Tsun-Yi Yang, Yi-Ting Chen, Yen-Yu Lin, Yung-Yu Chuang. Computer Vision and Pattern Recognition, 2019

299 citations recorded by the source corpus.

This paper proposes a method for head pose estimation from a single image. Previous methods often predict head poses through landmark or depth estimation and would require more computation than necessary. Our method is based on regression and feature aggregation. For having a compact model, we employ the soft stagewise regression scheme. Existing feature aggregation methods treat inputs as a bag of features and thus ignore their spatial relationship in a feature map. We propose to learn a fine-grained structure mapping for spatially grouping features before aggregation. The fine-grained structure provides part-based information and pooled values. By utilizing learnable and non-learnable importance over the spatial location, different model variants can be generated and form a complementary ensemble. Experiments show that our method outperforms the state-of-the-art methods including both the landmark-free ones and the ones based on landmark or depth estimation. With only a single RGB frame as input, our method even outperforms methods utilizing multi-modality information (RGB-D, RGB-Time) on estimating the yaw angle. Furthermore, the memory overhead of our model is 100 times smaller than those of previous methods.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/30732c0146c69306a4b62bb0ce6564d7ea4d1c71

How researchers see this paper

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

In Computer Vision and Pattern Recognition

Citations in the corpus

Deepfacelab: Integrated, flexible and extensible face-swapping framework (2020, methodology): “To measure the accuracy of the pose, we calculate the Euclidean distance between the Euler angles (extracted through FSA-Net [23]) of I t and I output .”
Energy-Based Models for Deep Probabilistic Regression (2020, methodology): “[59] Direct Gaussian Laplace Softmax (CE, L(2)) Softmax (CE, L(2), Var)”
LARGE: Latent-Based Regression through GAN Semantics (2021, methodology): “In (a) we compare to GHFeat [73], FSA [75] and SSV [34] on the CelebA-HQ dataset [24].”
How to Train Your Energy-Based Model for Regression (2020, background): “Regression is a fundamental problem within computer vision with many additional applications [25, 42, 45, 54, 56], which all would benefit from such accurate predictions.”