Optimisation Algorithms for Hand Posture Estimation
暫譯: 手勢姿態估計的優化演算法

Saremi, Shahrzad, Mirjalili, Seyedali

Optimisation Algorithms for Hand Posture Estimation
  • 出版商: Springer
  • 出版日期: 2019-09-04
  • 售價: $4,890
  • 貴賓價: 9.5 折 $4,646
  • 語言: 英文
  • 頁數: 205
  • 裝訂: Hardcover - also called cloth, retail trade, or trade
  • ISBN: 9811397562
  • ISBN-13: 9789811397561
  • 相關分類: Algorithms-data-structures

    • 海外代購書籍(需單獨結帳)

相關主題

商品描述

This book reviews the literature on hand posture estimation using generative methods, identifying the current gaps, such as sensitivity to hand shapes, sensitivity to a good initial posture, difficult hand posture recovery in cases of loss in tracking, and lack of addressing multiple objectives to maximize accuracy and minimize computational cost. To fill these gaps, it proposes a new 3D hand model that combines the best features of the current 3D hand models in the literature. It also discusses the development of a hand shape optimization technique. To find the global optimum for the single-objective problem formulated, it improves and applies particle swarm optimization (PSO), one of the most highly regarded optimization algorithms and one that is used successfully in both science and industry. After formulating the problem, multi-objective particle swarm optimization (MOPSO) is employed to estimate the Pareto optimal front as the solution for this bi-objective problem. The book also demonstrates the effectiveness of the improved PSO in hand posture recovery in cases of tracking loss. Lastly, the book examines the formulation of hand posture estimation as a bi-objective problem for the first time.

The case studies included feature 50 hand postures extracted from five standard datasets, and were used to benchmark the proposed 3D hand model, hand shape optimization, and hand posture recovery.

商品描述(中文翻譯)

本書回顧了使用生成方法進行手部姿勢估計的文獻，識別出當前的不足之處，例如對手部形狀的敏感性、對良好初始姿勢的敏感性、在追蹤丟失的情況下難以恢復手部姿勢，以及缺乏針對多個目標以最大化準確性和最小化計算成本的考量。為了填補這些空白，本書提出了一種新的3D手部模型，結合了文獻中現有3D手部模型的最佳特徵。它還討論了一種手部形狀優化技術的開發。為了找到所制定的單目標問題的全局最優解，本書改進並應用粒子群優化（Particle Swarm Optimization, PSO），這是一種備受推崇的優化算法，並在科學和工業中成功應用。在問題制定後，採用了多目標粒子群優化（Multi-Objective Particle Swarm Optimization, MOPSO）來估計帕累托最優前沿，作為此雙目標問題的解決方案。本書還展示了改進的PSO在追蹤丟失情況下手部姿勢恢復的有效性。最後，本書首次將手部姿勢估計的問題表述為雙目標問題。

所包含的案例研究特徵為從五個標準數據集中提取的50種手部姿勢，並用於基準測試所提出的3D手部模型、手部形狀優化和手部姿勢恢復。

作者簡介

 

Dr. Shahrzad Saremi is a lecturer at Griffith College, Griffith University, Australia. She received her BA in Information Technology from the Malaysian Multi Media University and M.Sc in Interaction Design from the University of Queensland. Dr. Saremi has published more than 20 articles in high-impact journals. Her main research interests include machine learning, optimization, human-computer interaction, augmented reality and gesture detection.

Dr. Seyedali Mirjalili is a lecturer at Griffith College, Griffith University and internationally recognized for his advances in nature-inspired Artificial Intelligence (AI) techniques. He is the author of five books, 100 journal articles, 20 conference papers, and 20 book chapters. With over 10000 citations and H-index of 40, he is one of the most influential AI researchers in the world. From Google Scholar metrics, he is globally the 3rd most cited researcher in Engineering Optimisation and Robust Optimisation using AI techniques. He has been the keynote speaker of several international conferences and is serving as an associate editor of top AI journals including Applied Soft Computing, Applied Intelligence, IEEE Access, Advances in Engineering Software, and Applied Intelligence.

 

作者簡介(中文翻譯)

 

Dr. Shahrzad Saremi 是澳洲格里菲斯大學格里菲斯學院的講師。她在馬來西亞多媒體大學獲得資訊科技學士學位，並在昆士蘭大學獲得互動設計碩士學位。Saremi 博士已在高影響力期刊上發表了超過 20 篇文章。她的主要研究興趣包括機器學習、優化、人機互動、擴增實境和手勢檢測。

Dr. Seyedali Mirjalili 是格里菲斯大學格里菲斯學院的講師，因其在自然啟發的人工智慧 (AI) 技術方面的進展而享有國際聲譽。他是五本書籍、100 篇期刊文章、20 篇會議論文和 20 篇書章的作者。Mirjalili 博士擁有超過 10000 次引用和 40 的 H 指數，是全球最具影響力的 AI 研究者之一。根據 Google Scholar 的指標，他在使用 AI 技術的工程優化和穩健優化領域中，全球引用次數排名第三。他曾擔任多個國際會議的主題演講者，並擔任多本頂尖 AI 期刊的副編輯，包括《應用軟體計算》、《應用智慧》、《IEEE Access》、《工程軟體進展》和《應用智慧》。

 

類似商品