ptcloud: bib

Author: berak

modules/ptcloud/README.md

@@ -4,14 +4,15 @@
 Point Cloud Module, Object Fitting API
 =======================================
 
-Try to segement geometric prmitives like planes, spheres and cylinders from a 3d point cloud
+Try to segment geometric primitives like planes, spheres and cylinders from a 3d point cloud
 
 2 alternative ransac strategies are implemented here:
 
 - plain ransac loop:
   + generate random minimal hypothesis
   + find inliers for the model,
     - bail out if at 1/4 of the data it does not have more than 1/8 inliers of the current best model
+    - if sprt is used, bail out if the probability of finding inliers goes low
   + if this model is the current best one
     - best model = current
     - update stopping criterion (optional SPRT)

modules/ptcloud/doc/ptcloud.bib

@@ -0,0 +1,24 @@
+@article{FB81
+  title={Random sample consensus:  A paradigm for model fitting with applications to image analysis and automated cartography},
+  author={Fischler, M. A. and Bolles, R.},
+  journal={Communications of the ACM},
+  year={1981},
+  volume={24},
+  number={2},
+  pages={381-395}
+}
+
+@inproceedings{NI03
+  title={Preemptive RANSAC for live structure and motion estimation},
+  author={Nister, D.},
+  booktitle={International Conference on Computer Vision (ICCV)},
+  year={2003}
+}
+
+@article{RA13
+  title={USAC: A Universal Framework for Random Sample Consensus},
+  author={Raguram, Rahul and Chum,Ondrej and Pollefeys, Marc and Matas,Jiri and Frahm, Jan-Michael},
+  journal={IEEE Transactions on Software Engineering},
+  year={2013},
+  organization={IEEE}
+}