Hi everyone, I am Ankit Agrawal, a 3rd year student enrolled in a 5 year Dual Degree Program(B.Tech + Masters) in Electrical Engineering at IIT Bombay(where SciPy India meet was held in Dec 2012). My Masters specialization is in Communication and Signal Processing. The relevant courses that I have taken in the past are Linear Algebra, Image Processing, Computer Vision and Machine Learning. Previously, I have used MATLAB, OpenCV and PointCloud Library for Image Processing and Computer Vision projects. Some of my projects include : 1) Mini Project *1*<http://home.iitb.ac.in/~aaaagrawal/projects/ip_project1.pdf>and *2 <http://home.iitb.ac.in/~aaaagrawal/projects/ip_project2.pdf>* based on paper 'Undersampled Radial MRI with Multiple Coils. Iterative Image Reconstruction Using a Total Variation Constraint<http://www-mrsrl.stanford.edu/studygroup/2/Files/Block_2007_Undersampled.pdf> ' 2) Parallel Tracking and Mapping<https://github.com/ankit-maverick/ComputerVisionProject>in 3D using Kinect(WIP) based on paper 'Parallel Tracking and mapping for Small AR Workspaces<http://www.robots.ox.ac.uk/~lav/Papers/klein_murray_ismar2007/klein_murray_i...> ' 3) Machine Learning competitions on Kaggle<https://www.kaggle.com/users/43981/ankit-agrawal#profile-results>(Yes, I have fallen victim to overfitting here.) I just got finished with my end-terms yesterday and hence was a bit under the hood during the last week and slow in pushing commits. I appreciate the patience of the community members who have been reviewing my Pull Requests. I have resumed working on them<https://github.com/scikit-image/scikit-image/pulls/ankit-maverick> . I want to discuss some of the ideas more extensively for building a strong GSoC proposal as the ideas page currently does not list much description of them. Idea # 1: In the feature detection idea, binary features like BRIEF<http://cvlabwww.epfl.ch/~lepetit/papers/calonder_eccv10.pdf>, FREAK <http://infoscience.epfl.ch/record/175537/files/2069.pdf> and BRISK<http://www.robots.ox.ac.uk/~vgg/rg/papers/brisk.pdf>; STAR corners, SURF<http://www.sciencedirect.com/science/article/pii/S1077314207001555>and SIFT <http://www.cs.ubc.ca/~lowe/papers/ijcv04.pdf> features need to be implemented(Currently, only feature readers are available for SIFT and SURF) . Over the span of 12 weeks, I think 4 of these can implemented giving 3 weeks of time to each. If this sounds a bit ambitious, 3 features can be completed with 4 for weeks for each one. I think a framework for handling features and descriptors like the opencv 2D features framework<http://docs.opencv.org/modules/features2d/doc/features2d.html>might be something that skimage should have in future considering the long-term. Idea # 2: I think that the Blind Deconvolution<http://people.csail.mit.edu/sparis/publi/2011/cvpr_radon/Cho_11_Blur_Kernel_Estimation.pdf>and Face Detection(@Stefan : Can you please confirm whether you meant Detection instead of Recognition in the idea listed on the idea's page<https://github.com/scikit-image/scikit-image/wiki/GSoC-2013>?) both can be completed comfortably in 12 weeks by me as I have worked on a project involving Radon transform(MiniProject 1<http://home.iitb.ac.in/~aaaagrawal/projects/ip_project1.pdf>and 2 <http://home.iitb.ac.in/~aaaagrawal/projects/ip_project2.pdf>) and have also done a course in Machine Learning. Currently, I don't know of any Non-patent-encumbered face detection algorithm in particular, but I am confident that there are effective methods other than Viola-Jones. A good introduction to Face detection and Recognition is in these slides<http://vision.stanford.edu/teaching/cs231a_autumn1112/lecture/lecture2_face_recognition_cs231a.pdf>by Stanford's Prof. Fei Fei Li. Beatka has shown interest in Graph-Cut as well as Blind Convolution, but I think it would be better if someone tackles all the segmentation algorithms together since the segmentation part of skimage needs many additions. There was a lot of interest for the segmentation idea in the beginning, but if no one is taking it, I would be happy to have that idea as my proposal. Idea # 3: What is needed to be implemented in Video Manipulation Framework idea? My guess is that interfacing the current Video streaming functionality(input either from webcam or files) with the currently implemented algorithms which would also lead to development of Computer Vision algorithms in the future in skimage. Can you please elaborate more on this idea, if possible in a modular way? I request all the community members to comment their views on these ideas soon as I plan to have my proposal ready by 30th. Thank you. Regards, Ankit Agrawal, Communication and Signal Processing, IIT Bombay.