Redirigiendo al acceso original de articulo en 18 segundos...
Inicio  /  Applied Sciences  /  Vol: 12 Par: 17 (2022)  /  Artículo
ARTÍCULO
TITULO

Benchmarking Deep Learning Models for Instance Segmentation

Sunguk Jung    
Hyeonbeom Heo    
Sangheon Park    
Sung-Uk Jung and Kyungjae Lee    

Resumen

Instance segmentation has gained attention in various computer vision fields, such as autonomous driving, drone control, and sports analysis. Recently, many successful models have been developed, which can be classified into two categories: accuracy- and speed-focused. Accuracy and inference time are important for real-time applications of this task. However, these models just present inference time measured on different hardware, which makes their comparison difficult. This study is the first to evaluate and compare the performances of state-of-the-art instance segmentation models by focusing on their inference time in a fixed experimental environment. For precise comparison, the test hardware and environment should be identical; hence, we present the accuracy and speed of the models in a fixed hardware environment for quantitative and qualitative analyses. Although speed-focused models run in real-time on high-end GPUs, there is a trade-off between speed and accuracy when the computing power is insufficient. The experimental results show that a feature pyramid network structure may be considered when designing a real-time model, and a balance between the speed and accuracy must be achieved for real-time application.

 Artículos similares

       
 
Eberechi Ichi, Faezeh Jafari and Sattar Dorafshan    
Annotated datasets play a significant role in developing advanced Artificial Intelligence (AI) models that can detect bridge structure defects autonomously. Most defect datasets contain visual images of surface defects; however, subsurface defect data su... ver más
Revista: Infrastructures

 
Nicola Convertini, Vincenzo Dentamaro, Donato Impedovo, Giuseppe Pirlo and Lucia Sarcinella    
This benchmarking study aims to examine and discuss the current state-of-the-art techniques for in-video violence detection, and also provide benchmarking results as a reference for the future accuracy baseline of violence detection systems. In this pape... ver más
Revista: Information

 
Ibrahim Furkan Ince    
Human activity recognition (HAR) has been an active area in computer vision with a broad range of applications, such as education, security surveillance, and healthcare. HAR is a general time series classification problem. LSTMs are widely used for time ... ver más
Revista: Applied Sciences

 
MohammedAli Mudhsh and Rolla Almodfer    
The traditional algorithms for recognizing handwritten alphanumeric characters are dependent on hand-designed features. In recent days, deep learning techniques have brought about new breakthrough technology for pattern recognition applications, especial... ver más
Revista: Information