paper-with-me

Papers

PSP-HDRI$+$: A Synthetic Dataset Generator for Pre-Training of Human-Centric Computer Vision Models

2022-07-11 · Salehe Erfanian Ebadi, Saurav Dhakad, Sanjay Vishwakarma, Chunpu Wang, You-Cyuan Jhang, Maciek Chociej, Adam Crespi, Alex Thaman, Sujoy Ganguly

We introduce a new synthetic data generator PSP-HDRI$+$ that proves to be a superior pre-training alternative to ImageNet and other large-scale synthetic data counterparts. We demonstrate that pre-training with our synthetic data will yield a more general model that performs better than alternatives even when tested on out-of-distribution (OOD) sets. Furthermore, using ablation studies guided by person keypoint estimation metrics with an off-the-shelf model architecture, we show how to manipulate our synthetic data generator to further improve model performance.

📄 PDF Abstract BibTeX arXiv:2207.05025

Code (1)

unity-technologies/peoplesanspeople 공식 구현

Tasks

Keypoint Estimation

Similar Papers 제목 키워드 기반

AmodalSynthDrive: A Synthetic Amodal Perception Dataset for Autonomous Driving

2023-09-12 · Ahmed Rida Sekkat, Rohit Mohan, Oliver Sawade, Elmar Matthes 외

Unlike humans, who can effortlessly estimate the entirety of objects even when partially occluded, modern computer vision algorithms still find this aspect extremely challenging. Leveraging this amodal perception for aut…

Autonomous DrivingBenchmarkingDepth EstimationScene Understanding

HDR Imaging for Dynamic Scenes with Events

2024-04-04 · Li Xiaopeng, Zeng Zhaoyuan, Fan Cien, Zhao Chen 외

High dynamic range imaging (HDRI) for real-world dynamic scenes is challenging because moving objects may lead to hybrid degradation of low dynamic range and motion blur. Existing event-based approaches only focus on a s…

DeblurringSelf-Supervised Learning

TorchDriveEnv: A Reinforcement Learning Benchmark for Autonomous Driving with Reactive, Realistic, and Diverse Non-Playable Characters

2024-05-07 · Jonathan Wilder Lavington, Ke Zhang, Vasileios Lioutas, Matthew Niedoba 외

The training, testing, and deployment, of autonomous vehicles requires realistic and efficient simulators. Moreover, because of the high variability between different problems presented in different autonomous systems, t…

Autonomous DrivingAutonomous Vehiclesreinforcement-learningReinforcement Learning

Fitting Spherical Gaussians to Dynamic HDRI Sequences

2024-12-09 · Pascal Clausen, Li Ma, Mingming He, Ahmet Levent Tasel 외

We present a technique for fitting high dynamic range illumination (HDRI) sequences using anisotropic spherical Gaussians (ASGs) while preserving temporal consistency in the compressed HDRI maps. Our approach begins with…

Deep Joint Demosaicing and High Dynamic Range Imaging within a Single Shot

2021-11-14 · Yilun Xu, Ziyang Liu, Xingming Wu, Weihai Chen 외

Spatially varying exposure (SVE) is a promising choice for high-dynamic-range (HDR) imaging (HDRI). The SVE-based HDRI, which is called single-shot HDRI, is an efficient solution to avoid ghosting artifacts. However, it …

Demosaicking