Learning Direct Optimization for scene understanding

PATTERN RECOGNITION

卷:105

Article

Romaszko, Lukasz¹  Williams, Christopher K., I^1,2  Winn, John³ 

[1] Univ Edinburgh, Sch Informat, 10 Crichton St, Edinburgh EH8 9AB, AB, Scotland

[2] Alan Turing Inst, 96 Euston Rd, London NW1 2DB, England

[3] Microsoft Res, 21 Stn Rd, Cambridge CB1 2FB, England

关键词: Computer vision;    Scene understanding;    3D Reconstruction;    Inverse graphics;    Object recognition;    Scene graph;    Analysis-by-synthesis;    Graphics;   

DOI  :  10.1016/j.patcog.2020.107369

来源: Elsevier

【 摘 要 】

We develop a Learning Direct Optimization (LiDO) method for the refinement of a latent variable model that describes input image x. Our goal is to explain a single image x with an interpretable 3D computer graphics model having scene graph latent variables z (such as object appearance, camera position). Given a current estimate of z we can render a prediction of the image g(z), which can be compared to the image x. The standard way to proceed is then to measure the error E(x, g(z)) between the two, and use an optimizer to minimize the error. However, it is unknown which error measure E would be most effective for simultaneously addressing issues such as misaligned objects, occlusions, textures, etc. In contrast, the LiDO approach trains a Prediction Network to predict an update directly to correct z, rather than minimizing the error with respect to z. Experiments show that LiDO converges rapidly as it does not need to perform a search on the error landscape, produces better solutions than error-based competitors, and is able to handle the mismatch between the data and the fitted scene model. We apply LiDO to a realistic synthetic dataset, and show that the method also transfers to work well with real images. (C) 2020 Elsevier Ltd. All rights reserved.

【 授权许可】

Free