2c684f4d174a1e4c2daa7d903d62bed6.ppt
- Количество слайдов: 8
Computational Rephotography Soonmin Bae Aseem Agarwala Frédo Durand
What is Rephotography The act of repeat photography of the same site – wikipedia Visualize changes over time n n Changes of a city or buildings Effects of erosion Need to decide n n n viewpoint season and time of day cameras and lenses
Newyork changing by Douglas Levere (New York City photographer) Custon house statues and New york produce exchange, Manhattan, 1936 Roadway to the Battery, Manhattan, 1938 and 1997 Native American museum statues and MTA headquarters, Manhattan, 1997
Urban Life though Two Lenses - 1890 and 2002
Goals Make rephotography easy n lead photographers to choosing a desired viewpoint matching that of a given image Visualize the difference between a given image and the current scene n n show to translate and rotate the camera Interact with users in a real time
Related work Depth (or shape) from stereo n n estimate a 3 D structure from two images using triangulation find pairs of corresponding image points and measure the difference to infer its depth Panoramic mosaic n n increase FOV use image-warping with homography (reprojection) Visual homing n n lead a robot to a desired positions and orientations Recover the epipolar geometry relating the current image taken by the robot and the target image
Background Homography n defines a geometric relation between two images Epipolar geometry find the geometric relation (homography) between two images of a single 3 D scene SIFT (scale-invariant feature transform) [Lowe 04] n extract distinctive features from images n the features are invariant to image scale, rotation, and partially invariant to changing viewpoints, change in illumination n SURF (speeded up robust features) n
Our Approach find and match distinctive features of a given image and the scene n using Computer vision techniques (e. g. , SIFT and SURF) compute the amount of translation n using epipolar geometry interact with users in a real time and lead to a desired viewpoint n using visualization resolve rotation n n using image warping with homographies or lead users to rotating the camera
2c684f4d174a1e4c2daa7d903d62bed6.ppt