Projects for Digital Video Processing (EE392J)

Possible Projects

• Digital Processing of Analog Television (Software TV)  Digitize an analog television (NTSC) signal and use a computer (e.g. Matlab) to perform the "decoding" to produce the video frames. Develop and apply comb filters (to separate luminance from chrominance), deinterlacing algorithm (to convert from interlaced to progressive) and other algorithms to improve the visual quality.
• Frame-Rate Conversion Develop an algorithm for motion-compensated interpolation. Important frame-rate conversions include 10 f/s to 20 f/s, 24 f/s to 30 f/s (film to TV), and 50 f/s to 60 f/s (PAL to NTSC). 
• Superresolution Develop an algorithm that uses a number of low-resolution video frames to create a high resolution still frame. Develop an algorithm that takes low-resolution video and creates enhanced-resolution video. 
• Mosaicing Develop an algorithm that takes a number of low resolution frames of a scene to generate a single large panoramic of the scene. 
• Deinterlacing Develop an algorithm for interlaced to progressive conversion. Possible applications include display of interlaced TV on a progressive computer monitor, generation of a high quality frame from an interlaced video of a still scene (e.g. for printing a picture from TV), and standards-conversion such as 480-line interlaced to 480-line progressive or 1080-line interlaced to 720-line progressive (important for Digital TV). 
• Video Stabilization Handheld video cameras are often afflicted by unintentional camera movement.  Develop an algorithm that identifies this unintentional movement and compensates for it.
• Video Watermarking Develop an algorithm to place an invisible watermark in a video, where the existence of the watermark can be identified by an associated watermark detection algorithm.  In addition to being invisible and relatively easy to detect, the watermark should be difficult to remove.  For example, adding noise or encoding/decoding the video should not destroy the watermark.
• Video Fingerprinting Develop an algorithm that produces a fingerprint of an image or video where this fingerprint can be used to recognize the image or video.  The fingerprint should be robust to various operations (e.g. adding noise, encoding/decoding), and therefore accurate recognition would be possible even when the signal is processed via these operations.  Fingerprinting is an example of media identification by passive watermarking, where the media is unchanged, in contrast to the active watermarking above, where the media is changed.
• Segmentation Develop an algorithm to perform foreground/background segmentation of a scene. For example, separate the speaker in a video conference or news show. Develop an algorithm to identify the number of people in a scene or in a room. 
• Restoration and Noise Reduction Develop an algorithm to deblur a (blurred or out of focus) video signal and reduce other artifacts (e.g. motion picture film is often afflicted by scratches and salt-and-pepper noise). 
• Motion Estimation Compare different algorithms for performing motion estimation. Algorithms may include different methods for gradient-based and/or block-based estimation, single-layer versus multi-layer estimation. Comparison criteria could include closeness to the true motion in the scene, performance when used for compression (e.g. energy in MC-prediction error), and complexity. 
• Object-Tracking Develop an algorithm to track an object in a video scene. Examples including tracking a soccer ball in a soccer game, cars driving down a street, or people moving in a room. The algorithm may be fully automatic, or semi-automatic where the user initializes the algorithm by telling it where the object(s) is (are) in the first frame. 
• Stereo Processing Develop an algorithm(s) to perform disparity estimation. Evaluate your algorithm(s).  
• Content-Based Sampling Develop an algorithm to identify and extract key-frames from a video sequence so that those frames may be used, for example, to provide a summary of the video or to enable efficient searching of the video. 

Other Possible Projects 

Other video-related projects are also possible as long as they will be educational to the student and they can realistically be completed in the time available. Please contact the instructors about other project proposals.

Project Timetable:

• Proposal (1-2 pages), due shortly after midterm. 
• Report, due last day of class.  The report should contain no more than 10 pages of text per person (not per project); extra pages for images, figures, references is OK. 
• Brief oral presentation (15-20 minutes/project), during last day of class or final exam period. 

Project Planning

The project should be defined with two sets of goals. The first set of goals should be such that you are confident that they can realistically be achieved in the time available. The second set of goals should be more challenging and potentially you may be able to finish them, or maybe only a subset of them.  The instructors will help in designing these goals (based on the students' background and interests) and provide direction to appropriate literature outside the textbook.  The Additional References will also be helpful in designing and working on the final project.

Project Team

Each project may be performed by a single student or a group of two or three students. Teamwork is encouraged, however the proposed work for each individual should be identified in the proposal, and the actual contributions of each individual should be identified in the report. For example, if a two-person team is working on deinterlacing then each student can develop separate deinterlacing algorithms and they can compare the performance. 

Equipment Requirements 

The above projects have a variety of requirements in terms of equipment, e.g. video capture capability, stereo camera, etc. While we will try to assist with these whenever possible (e.g. video capture capability), please consider carefully the equipment requirements and what equipment you have access to when choosing a project.