Exploring Rare Pose in Human Pose Estimation

Hwang, Jee Youn; Yang, John; Kwak, Nojun

doi:10.1109/access.2020.3033531

Cited by 8 publications

(4 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since pose naturalness should be evaluated without considering frequency -understanding that rare poses can be natural and common ones peculiar -we focused on measuring peculiarity independently of frequency. Drawing inspiration from Hwang et al (Hwang, Yang, and Kwak 2020), we developed our metric using clustering applied to MS COCO dataset (Lin et al 2014), which is rich in real human images. For each of N persons in MS COCO dataset, we performed sample-wise min-max normalization on their poses as:…”

Section: Measuring Peculiarity In Generated Imagesmentioning

confidence: 99%

Rethinking Peculiar Images by Diffusion Models: Revealing Local Minima’s Role

Jang,

Youn,

Jeon

et al. 2024

AAAI

View full text Add to dashboard Cite

Recent significant advancements in diffusion models have revolutionized image generation, enabling the synthesis of highly realistic images with text-based guidance. These breakthroughs have paved the way for constructing datasets via generative artificial intelligence (AI), offering immense potential for various applications. However, two critical challenges hinder the widespread adoption of synthesized data: computational cost and the generation of peculiar images. While computational costs have improved through various approaches, the issue of peculiar image generation remains relatively unexplored. Existing solutions rely on heuristics, extra training, or AI-based post-processing to mitigate this problem. In this paper, we present a novel approach to address both issues simultaneously. We establish that both gradient descent and diffusion sampling are specific cases of the generalized expectation maximization algorithm. We hypothesize and empirically demonstrate that peculiar image generation is akin to the local minima problem in optimization. Inspired by optimization techniques, we apply naive momentum and positive-negative momentum to diffusion sampling. Last, we propose new metrics to evaluate the peculiarity. Experimental results show momentum effectively prevents peculiar image generation without extra computation.

show abstract

Section: Measuring Peculiarity In Generated Imagesmentioning

confidence: 99%

Rethinking Peculiar Images by Diffusion Models: Revealing Local Minima’s Role

Jang,

Youn,

Jeon

et al. 2024

AAAI

View full text Add to dashboard Cite

show abstract

“…In [11], Hwang et al, present an architecture to address the human pose estimation problem where data mismatch between various postures. They explore poses of unusual to identify without additional learning of a rare pose and used a simple K-means algorithm of clustering.…”

Section: B Multi Person Pose Estimationmentioning

confidence: 99%

“…Besides, the runtime is directly correlated with the population of the image [7], [8], [9]. Conversely, bottom-up strategies are desirable because they give stability to early investment and have the capacity to dissociate runtime challenge based on the number of people in the picture [3], [4], [10], [11].…”

Section: Introductionmentioning

confidence: 99%

Multi-Person Pose Estimation Using Group-Based Convolutional Neural Network Model

et al. 2023

View full text Add to dashboard Cite

Human pose estimation has drawn extensive attention recently and there has been significant progress on it due to the rising popularity of convolutional neural networks (CNN). However, existing stateof-the-art approaches suffer from occlusion, complicated backgrounds, and substantial position fluctuations because of disregarding the human body form. Human parsing is a very pertinent activity that can provide crucial semantic data about bodily parts for position estimation. To overcome the aforesaid limitations, this paper introduces a human pose estimation method using a group-based convolutional neural network model. The proposed method adopts a bottom-up parsing strategy that yields features to extract skeletal key points in the human body. Moreover, it creates a grouping of anatomical key points for individuals by utilizing the non-parametric description for the key point association vector field. Experimental results indicate that the proposed method provides superior performance than the state-of-the-art algorithms in terms of accuracy. In addition, it optimizes its output and detects occluded as well as invisible key points by incorporating feature representation. The proposed method surpasses the recent methods, achieving 93% of the mean average accuracy.

show abstract

“…Ye et al [12] used a hybrid method to estimate gestures in a multi-stage and hierarchical manner, which has good robustness. Hwang, J et al [13] used a simple Kmeans clustering algorithm to explore unusual poses that are rare which occupy a small portion in a pose dataset. B Artacho et al [14] proposed OmniPose, a single-pass, endto-end trainable framework, that achieves state-of-the-art results for multi-person pose estimation.…”

Section: Introductionmentioning

confidence: 99%

Lightweight Gesture Pose Estimation Based on CPM Algorithm

Wang

Yang

Wang

et al. 2022

Modern Management Based on Big Data III

View full text Add to dashboard Cite

With the experiential enhancement of artificial intelligence products, gesture estimation, as a classic computer vision task, has a wide range of application scenarios. Aiming at the current network model that needs to be lightweight in mobile smart products, this paper designs a lightweight gesture pose estimation model based on the CPM (Convolutional Pose Machine) multi-stage human pose estimation network. A comparative experiment based on the RHD open-source data set was conducted to compare and analyze the lightweight CPM gesture estimation model while ensuring accuracy while effectively reducing the amount of model parameters, which provides a basis for the development of real-time mobile terminal gesture pose estimation.

show abstract

Exploring Rare Pose in Human Pose Estimation

Cited by 8 publications

References 28 publications

Rethinking Peculiar Images by Diffusion Models: Revealing Local Minima’s Role

Rethinking Peculiar Images by Diffusion Models: Revealing Local Minima’s Role

Multi-Person Pose Estimation Using Group-Based Convolutional Neural Network Model

Lightweight Gesture Pose Estimation Based on CPM Algorithm

Contact Info

Product

Resources

About