Modeling and Rendering Architecture from Photographs:
A Hybrid Geometry- and Image-Based Approach

Images and Movies

Images

Fig. 1. Schematic comparison of geometry-based and image-based modeling/rendering systems, and our hybrid approach

Fig. 2. Image viewer showing marked features and model viewer showing recovered model images from the photogrammetric modeling system

Fig. 3. Reprojected model edges and rendered view of clock tower

Fig. 9. Three of twelve images used to reconstruct a high school building, with marked features in green

Fig. 10. Recovered model edges reprojected through recovered camera positions into same three images

Fig. 11. Three views of the recovered high school model and camera positions

Fig. 12. Process of projecting images onto the recovered model with view-dependent texture-mapping

Fig. 14. Examination of view-dependent texture-mapping

Fig. 16. Key, Warped-Offset, and Offset images used in model-based stereo algorithm

Fig. 18. Disparity map computed by model-based stereo algorithm

Fig. 19. Rendered views of recovered chapel facade model, which are full-size images of frames 68, 0, and 290 of movie6.mov

Movies (Apple QuickTime format)

Four images projected onto recovered high school model

Twelve images projected onto recovered high school model with view-dependent texture-mapping

Same as movie2.mov, with obstructions (signs, trees) masked out of the original images

Fly-around of chapel facade with traditional texture-mapping

Fly-around of chapel facade with view-dependent texture-mapping using four images

Fly-around of chapel facade with geometric detail recovered from model-based stereo and composited with view-dependent texture-mapping using the same four images


Paul Debevec (debevec@cs.berkeley.edu)