Upload
zalika
View
92
Download
0
Tags:
Embed Size (px)
DESCRIPTION
IBMR Assignment 1. Stitching Photo Mosaics. What is Photo Mosaic?. Stitching photos to construct a wild-view scene. Part1: CORNER DETECTION Part2: PERSPECTIVE MAPPING and MOSAICING Handout after Part2 Finished. PERSPECTIVE MAPPING and MOSAICING. Requirement. - PowerPoint PPT Presentation
Citation preview
IBMR Assignment 1Stitching Photo Mosaics
What is Photo Mosaic?• Stitching photos to construct a wild-view scene.• Part1: CORNER DETECTION• Part2: PERSPECTIVE MAPPING and
MOSAICING• Handout after Part2 Finished
PERSPECTIVE MAPPING and MOSAICING
Requirement• Read n>2 images, and create an image mosaic
by registering, projective warping, resampling, and compositing them.
• (bonus) multiband blending, SIFT ,panorama or other methods mentioned in class.
Just overlapping
Steps• Shoot the Pictures• Recover Homographies• Warp the Images/ Image Rectification• Gain Compensation• Blend the images into a mosaic
Shoot the Pictures• You may use the photos on the webpage, but
shoot your own photos and mosaic them will get bonus credit.
• Shoot photos as:
• Overlap the fields of view significantly. 40% to 70% overlap is recommended.
Recover Homographies• Construct a linear system as: p’=Hp, where p’ and
p are correspondence points.• Follow the Lecture 8 page 6~9. You may try Affine
mappings(DOF=6) or Projective mappings(DOF=8).
• Solve Ax=0
Warp the Images/Image Rectification• Source scanning(forward mapping) or destination
scanning(inverse mapping).• You will need to avoid aliasing when resampling
the image. • Be careful of the size of the resulting image.
Gain Compensation• Find the optimize gains of gi according to means
of overlapping regions between image pair i and j.
Blend the images into a mosaic• Linear blending by the weights:
where w(x) varies linearly from 1 at the centre of
the image to 0 at the edge.
• Multi-band blending (bonus):
A B𝐼 (𝑥 , 𝑦 )=
𝐼𝑎∗ (𝑊 −𝐷𝑎)+𝐼𝑏∗(𝑊 −𝐷𝑏)(𝑊 −𝐷𝑎 )+(𝑊 −𝐷𝑏)
Multi-band blending• Band 1 scale 0 to σ
• Band 2 scale σ to 2σ
• Band 3 lower than 2σ
Support• Your own project1a code.• A C called matlab library.• to calculate inverse matrix , SVD or etc.
Result
Grading• Basic: 75%• Harris Corner Detection + KNN (Hw1a)• RANSAC• Projection Mapping / Affine Mapping• Image Warping
• Bonus:• Non-Maximum Suppression 5%• KD Tree 5%• SIFT 15%• Gain Compensation 10%• Linear Blending 5%• Multi Blending10%• Stitching your own photos 5%• Others
Deadline• 11/22 11:59:59pm• Upload your program & report to:• host : caig.cs.nctu.edu.tw• port : 30021• username : IBMR10• password : IBMR10
• and create your own folder with your ID.