A Confidence Measure for Assessing Optical Flow Accuracy in the Absence of Ground Truth

被引：0
|
作者：
Marquez, Patricia [1 ]
Gil, Debora [1 ]
Hernandez-Sabate, Aura [1 ]
机构：
[1] Comp Vis Ctr, Barcelona, Spain
来源：
2011 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS (ICCV WORKSHOPS) | 2011年
关键词：
D O I：
暂无
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
Optical flow is a valuable tool for motion analysis in autonomous navigation systems. A reliable application requires determining the accuracy of the computed optical flow. This is a main challenge given the absence of ground truth in real world sequences. This paper introduces a measure of optical flow accuracy for Lucas-Kanade based flows in terms of the numerical stability of the data-term. We call this measure optical flow condition number. A statistical analysis over ground-truth data show a good statistical correlation between the condition number and optical flow error. Experiments on driving sequences illustrate its potential for autonomous navigation systems.
引用
收藏
页数：8
相关论文
共 50 条
  • [1] Learning a Confidence Measure for Optical Flow
    Mac Aodha, Oisin
    Humayun, Ahmad
    Pollefeys, Marc
    Brostow, Gabriel J.
    IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2013, 35 (05) : 1107 - 1120
  • [2] Measuring Relative Accuracy of Malware Detectors in the Absence of Ground Truth
    Charlton, John
    Du, Pang
    Cho, Jin-Hee
    Xu, Shouhuai
    2018 IEEE MILITARY COMMUNICATIONS CONFERENCE (MILCOM 2018), 2018, : 450 - 455
  • [3] Bootstrap optical flow confidence and uncertainty measure
    Kybic, Jan
    Nieuwenhuis, Claudia
    COMPUTER VISION AND IMAGE UNDERSTANDING, 2011, 115 (10) : 1449 - 1462
  • [4] Reverse Classification Accuracy: Predicting Segmentation Performance in the Absence of Ground Truth
    Valindria, Vanya V.
    Lavdas, Ioannis
    Bai, Wenjia
    Kamnitsas, Konstantinos
    Aboagye, Eric O.
    Rockall, Andrea G.
    Rueckert, Daniel
    Glocker, Ben
    IEEE TRANSACTIONS ON MEDICAL IMAGING, 2017, 36 (08) : 1597 - 1606
  • [5] Assessing the Quality of Spatio-Textual Datasets in the Absence of Ground Truth
    Ge, Mouzhi
    Chondrogiannis, Theodoros
    NEW TRENDS IN DATABASES AND INFORMATION SYSTEMS, ADBIS 2017, 2017, 767 : 12 - 20
  • [6] Assessing the accuracy of non-rigid registration with and without ground truth
    Schestowitz, R.
    Twining, C. J.
    Cootes, T.
    Petrovi, V.
    Taylor, C. J.
    Crum, W. R.
    2006 3RD IEEE INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING: MACRO TO NANO, VOLS 1-3, 2006, : 836 - +
  • [7] A heuristic measure for detecting influence of lossy JP2 compression on Optical Character Recognition in the absence of ground truth
    Schlarb, Sven
    Neudecker, Clemens
    ARCHIVING 2012: PRESERVATION STRATEGIES AND IMAGING TECHNOLOGIES FOR CULTURAL HERITAGE INSTITUTIONS AND MEMORY ORGANIZATIONS, 2012, : 250 - +
  • [8] A Method for Assessing Ground-Truth Accuracy of the 5DCT Technique
    Dou, Tai H.
    Thomas, David H.
    O'Connell, Dylan P.
    Lamb, James M.
    Lee, Percy
    Low, Daniel A.
    INTERNATIONAL JOURNAL OF RADIATION ONCOLOGY BIOLOGY PHYSICS, 2015, 93 (04): : 925 - 933
  • [9] Evaluation of the Capabilities of Confidence Measures for Assessing Optical Flow Quality
    Marquez-Valle, Patricia
    Gil, Debora
    Hernandez-Sabate, Aura
    2013 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS (ICCVW), 2013, : 624 - 631
  • [10] A ground truth correspondence measure for benchmarking
    Karlsson, Johan
    Ericsson, Anders
    18TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOL 3, PROCEEDINGS, 2006, : 568 - +
12345