paper-with-me

Papers

Efficient, Self-Supervised Human Pose Estimation with Inductive Prior Tuning

2023-11-06 · Nobline Yoo, Olga Russakovsky

The goal of 2D human pose estimation (HPE) is to localize anatomical landmarks, given an image of a person in a pose. SOTA techniques make use of thousands of labeled figures (finetuning transformers or training deep CNNs), acquired using labor-intensive crowdsourcing. On the other hand, self-supervised methods re-frame the HPE task as a reconstruction problem, enabling them to leverage the vast amount of unlabeled visual data, though at the present cost of accuracy. In this work, we explore ways to improve self-supervised HPE. We (1) analyze the relationship between reconstruction quality and pose estimation accuracy, (2) develop a model pipeline that outperforms the baseline which inspired our work, using less than one-third the amount of training data, and (3) offer a new metric suitable for self-supervised settings that measures the consistency of predicted body part length proportions. We show that a combination of well-engineered reconstruction losses and inductive priors can help coordinate pose learning alongside reconstruction in a self-supervised paradigm.

📄 PDF Abstract BibTeX arXiv:2311.02815

Code (1)

princetonvisualai/hpe-inductive-prior-tuning 공식 구현 pytorch

Tasks

2D Human Pose EstimationPose Estimation

Similar Papers 제목 키워드 기반

You Don't Need Strong Assumptions: Visual Representation Learning via Temporal Differences

2026-06-14 · Ninad Daithankar, Alexi Gladstone, Yann LeCun, Heng Ji arxiv

Progress in AI has largely been driven by methods that assume less. As compute and data increase, approaches with weaker inductive biases generally outperform those with stronger assumptions. This is particularly charact…

Self-Supervised LearningRepresentation Learning

DEFLOW: Self-supervised 3D Motion Estimation of Debris Flow

2023-04-05 · Liyuan Zhu, Yuru Jia, Shengyu Huang, Nicholas Meyer 외

Existing work on scene flow estimation focuses on autonomous driving and mobile robotics, while automated solutions are lacking for motion in nature, such as that exhibited by debris flows. We propose DEFLOW, a model for…

Autonomous DrivingDepth EstimationMotion EstimationOptical Flow Estimation+2

3D Packing for Self-Supervised Monocular Depth Estimation

2019-05-06 · CVPR 2020 6 · Vitor Guizilini, Rares Ambrus, Sudeep Pillai, Allan Raventos 외

Although cameras are ubiquitous, robotic platforms typically rely on active sensors like LiDAR for direct 3D perception. In this work, we propose a novel self-supervised monocular depth estimation method combining geomet…

Depth EstimationInductive BiasMonocular Depth EstimationSelf-Driving Cars

Test-Time Personalization with a Transformer for Human Pose Estimation

2021-07-05 · NeurIPS 2021 12 · Yizhuo Li, Miao Hao, Zonglin Di, Nitesh B. Gundavarapu 외

We propose to personalize a human pose estimator given a set of test images of a person without using any manual annotations. While there is a significant advancement in human pose estimation, it is still very challengin…

Pose Estimation

Purrturbed but Stable: Human-Cat Invariant Representations Across CNNs, ViTs and Self-Supervised ViTs

2025-11-04 · Arya Shah, Vaibhav Tripathi arxiv

Cats and humans differ in ocular anatomy. Most notably, Felis Catus (domestic cats) have vertically elongated pupils linked to ambush predation; yet, how such specializations manifest in downstream visual representations…