Lightweight Super-Resolution Head for Human Pose Estimation
- URL: http://arxiv.org/abs/2307.16765v1
- Date: Mon, 31 Jul 2023 15:35:34 GMT
- Title: Lightweight Super-Resolution Head for Human Pose Estimation
- Authors: Haonan Wang, Jie Liu, Jie Tang, Gangshan Wu
- Abstract summary: Heatmap-based methods have become the mainstream method for pose estimation.
However, heatmap-based approaches suffer from significant quantization errors with downscale heatmaps.
We propose SRPose to reduce the quantization error and dependence on further post-processing.
- Score: 42.51588635059534
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Heatmap-based methods have become the mainstream method for pose estimation
due to their superior performance. However, heatmap-based approaches suffer
from significant quantization errors with downscale heatmaps, which result in
limited performance and the detrimental effects of intermediate supervision.
Previous heatmap-based methods relied heavily on additional post-processing to
mitigate quantization errors. Some heatmap-based approaches improve the
resolution of feature maps by using multiple costly upsampling layers to
improve localization precision. To solve the above issues, we creatively view
the backbone network as a degradation process and thus reformulate the heatmap
prediction as a Super-Resolution (SR) task. We first propose the SR head, which
predicts heatmaps with a spatial resolution higher than the input feature maps
(or even consistent with the input image) by super-resolution, to effectively
reduce the quantization error and the dependence on further post-processing.
Besides, we propose SRPose to gradually recover the HR heatmaps from LR
heatmaps and degraded features in a coarse-to-fine manner. To reduce the
training difficulty of HR heatmaps, SRPose applies SR heads to supervise the
intermediate features in each stage. In addition, the SR head is a lightweight
and generic head that applies to top-down and bottom-up methods. Extensive
experiments on the COCO, MPII, and CrowdPose datasets show that SRPose
outperforms the corresponding heatmap-based approaches. The code and models are
available at https://github.com/haonanwang0522/SRPose.
Related papers
- Learning Structure-Guided Diffusion Model for 2D Human Pose Estimation [71.24808323646167]
We propose textbfDiffusionPose, a new scheme for learning keypoints heatmaps by a neural network.
During training, the keypoints are diffused to random distribution by adding noises and the diffusion model learns to recover ground-truth heatmaps from noised heatmaps.
Experiments show the prowess of our scheme with improvements of 1.6, 1.2, and 1.2 mAP on widely-used COCO, CrowdPose, and AI Challenge datasets.
arXiv Detail & Related papers (2023-06-29T16:24:32Z) - Poseur: Direct Human Pose Regression with Transformers [119.79232258661995]
We propose a direct, regression-based approach to 2D human pose estimation from single images.
Our framework is end-to-end differentiable, and naturally learns to exploit the dependencies between keypoints.
Ours is the first regression-based approach to perform favorably compared to the best heatmap-based pose estimation methods.
arXiv Detail & Related papers (2022-01-19T04:31:57Z) - FasterPose: A Faster Simple Baseline for Human Pose Estimation [65.8413964785972]
We propose a design paradigm for cost-effective network with LR representation for efficient pose estimation, named FasterPose.
We study the training behavior of FasterPose, and formulate a novel regressive cross-entropy (RCE) loss function for accelerating the convergence.
Compared with the previously dominant network of pose estimation, our method reduces 58% of the FLOPs and simultaneously gains 1.3% improvement of accuracy.
arXiv Detail & Related papers (2021-07-07T13:39:08Z) - Improving Robustness for Pose Estimation via Stable Heatmap Regression [19.108116394510258]
A heatmap regression method is proposed to alleviate network vulnerability to small perturbations.
A maximum stability training loss is used to simplify the optimization difficulty.
The proposed method achieves a significant advance in robustness over state-of-the-art approaches on two benchmark datasets.
arXiv Detail & Related papers (2021-05-08T03:07:05Z) - Heatmap Regression via Randomized Rounding [105.75014893647538]
We propose a simple yet effective quantization system to address the sub-pixel localization problem.
The proposed system encodes the fractional part of numerical coordinates into the ground truth heatmap using a probabilistic approach during training.
arXiv Detail & Related papers (2020-09-01T04:54:22Z) - Attentive One-Dimensional Heatmap Regression for Facial Landmark
Detection and Tracking [73.35078496883125]
We propose a novel attentive one-dimensional heatmap regression method for facial landmark localization.
First, we predict two groups of 1D heatmaps to represent the marginal distributions of the x and y coordinates.
Second, a co-attention mechanism is adopted to model the inherent spatial patterns existing in x and y coordinates.
Third, based on the 1D heatmap structures, we propose a facial landmark detector capturing spatial patterns for landmark detection on an image.
arXiv Detail & Related papers (2020-04-05T06:51:22Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.