日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

物体検出arXiv:2203.10194

航空画像の物体検出に向けたYOLOv4の分析と適応

Analysis and Adaptation of YOLOv4 for Object Detection in Aerial Images

シェア:XThreadsFacebookLINEはてブBluesky

ドローン搭載を想定し、転移学習でYOLOv4をVisDroneデータセットに適応、精度と推論速度のバランスや解像度・学習回数の影響を評価した研究。

著者: Aryaman Singh Samyal, Akshatha K R, Soham Hans, Karunakar A K, Satish Shenoy B

分類: cs.CV, cs.AI, cs.LG, cs.RO

原文アブストラクト

The recent and rapid growth in Unmanned Aerial Vehicles (UAVs) deployment for various computer vision tasks has paved the path for numerous opportunities to make them more effective and valuable. Object detection in aerial images is challenging due to variations in appearance, pose, and scale. Autonomous aerial flight systems with their inherited limited memory and computational power demand accurate and computationally efficient detection algorithms for real-time applications. Our work shows the adaptation of the popular YOLOv4 framework for predicting the objects and their locations in aerial images with high accuracy and inference speed. We utilized transfer learning for faster convergence of the model on the VisDrone DET aerial object detection dataset. The trained model resulted in a mean average precision (mAP) of 45.64% with an inference speed reaching 8.7 FPS on the Tesla K80 GPU and was highly accurate in detecting truncated and occluded objects. We experimentally evaluated the impact of varying network resolution sizes and training epochs on the performance. A comparative study with several contemporary aerial object detectors proved that YOLOv4 performed better, implying a more suitable detection algorithm to incorporate on aerial platforms.

関連論文

  1. 1Boston Dynamics、車輪付きAtlasの研究試作機を紹介研究・技術
  2. 2newmoとKDDIが資本業務提携 2028年に堺市で自動運転タクシー商用化目指す提携・買収
  3. 3Google DeepMind、ロボット向けAIを発表新製品・発表
  4. 4NVIDIA、CoRLでGR00T 2とIsaac Lab 3.0を発表新製品・発表
  5. 5GXOとApptronik、Apolloの物流現場試験を拡大 最大1,000台の展開を構想量産・展開

PR・アフィリエイト広告

OZ GAMING(GPU搭載BTOゲーミングPC)

OZ GAMING(GPU搭載BTOゲーミングPC)の公式広告バナーOZ GAMINGで完成品PCを見る ↗
PR本紙発行元 EmplifAI