2013 IEEE International Conference on Computer Vision 2013
DOI: 10.1109/iccv.2013.94
|View full text |Cite
|
Sign up to set email alerts
|

Real-World Normal Map Capture for Nearly Flat Reflective Surfaces

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
7
0
1

Year Published

2015
2015
2020
2020

Publication Types

Select...
5
3
1

Relationship

0
9

Authors

Journals

citations
Cited by 12 publications
(8 citation statements)
references
References 31 publications
0
7
0
1
Order By: Relevance
“…While the ellipsoid fit is an approximation -since car windows can have small imperfections and normal variations [Jacquet et al 2013] -the resulting flow is sufficiently plausible in our test scenes, and allows real-time performance. See Fig.…”
Section: Ellipsoid Fitting For Car Windowsmentioning
confidence: 99%
“…While the ellipsoid fit is an approximation -since car windows can have small imperfections and normal variations [Jacquet et al 2013] -the resulting flow is sufficiently plausible in our test scenes, and allows real-time performance. See Fig.…”
Section: Ellipsoid Fitting For Car Windowsmentioning
confidence: 99%
“…Do tính chất và tầm quan trọng, có khá nhiều kỹ thuật được đề xuất phục vụ việc xây dựng bản đồ pháp tuyến bề mặt [4,5,8,7,11,16,20,22,24,26,27] tuy nhiên hầu hết các phương pháp có độ chính xác chưa cao hoặc phức tạp trong việc tổ chức và thu thập dữ liệu, hoặc không không phù hợp với các dạng bề mặt có hệ số phản xạ gương cao. Có thể phân các kỹ thuật ước lượng pháp tuyến thành ba hướng chính:…”
Section: Giới Thiệuunclassified
“…Other algorithms have also been proposed to recover 3D shapes from specular images. They employed either a known calibrated scene (Savarese et al, 2005; Tarini et al, 2005; Liu et al, 2013) or multiple images such as specular flows (Adato et al, 2010), motions of reflection correspondences (Sankaranarayanan et al, 2010), or line tracking (Jacquet et al, 2013). Although they are useful in some situations, they cannot recover 3D shapes from a single specular image in an unknown scene.…”
Section: Introductionmentioning
confidence: 99%