Neueste Forschungspaper wären zum Beispiel:
P. Viola and M. Jones, "Rapid Object Detection using a Boosted Cascade of Simple Features", 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'01), vol.1, pp. 511, 2001.
http://www.robots.ox.ac.uk/~cvrg/trinity2002/CVPR-2001.pdf
M. B. Blaschko and C. H. Lampert, "Learning to Localize Objects with Structured Output Regression", Computer Vision: ECCV 2008, 2-15.
http://www.kyb.mpg.de/publications/attachments/ECCV2008-Blaschko_5247[0].pdf
Torralba, A., Murphy, K. and Freeman, W., "Sharing Features: Efficient Boosting Procedures for Multiclass Object Detection", CVPR 2004
http://people.cs.ubc.ca/~murphyk/Papers/jointBoostCVPR04_camera.pdf
Pedro F. Felzenszwalb, Ross B. Girshick, David McAllester and Deva Ramanan, "Object Detection with Discriminatively Trained Part Based Models", PAMI 2009.
http://people.cs.uchicago.edu/~rbg/...-Felzenszwalb-Girshick-McAllester-Ramanan.pdf
Ansonsten ist "Sliding Windows" ein gutes Suchwort, weil das der derzeit meistgenutzte Ansatz ist. Für all das gute mathematische Kenntnisse und Grundlagen in Machine Learning allerdings vorausgesetzt.