Video interaction through finger tips

27
WELCOME

Transcript of Video interaction through finger tips

Page 1: Video interaction through finger tips

WELCOME

Page 2: Video interaction through finger tips

VIDEO INTERACTION THROUGH FINGER

TIPS

Coined By:

Nithin Prince JohnRoll No : 11M.Tech CSESBCE, Pattoor

Guided By:

Keerthi A S PillaiAssistant ProfessorDept of CSESBCE, Pattoor

Page 3: Video interaction through finger tips

CONTENTS

OBJECTIVE

INTRODUCTION

PROPOSED SYSTEM

MODULES OF THE SYSTEM

RESULT

CONCLUSION

5/23/2016 Computer Science & Engg 3

Page 4: Video interaction through finger tips

OBJECTIVE

Leverage existing content, turn them into transactional content marketing

5/23/2016 4Computer Science & Engg

Page 5: Video interaction through finger tips

“4 in 10 shoppers visited a store in person or online as a direct result of watching a video”

5/23/2016 5Computer Science & Engg

Page 6: Video interaction through finger tips

INTRODUCTION

Today online videos provide a passive user-experience for users

Interactive video enables users to take actions right within the video content

Interruption is the core issue

Developed a technique to blend interaction with any video, new or existing

5/23/2016 6Computer Science & Engg

Page 7: Video interaction through finger tips

Contd...

Interactivity makes videos more engaging and immersive

On the back end, items are tracked and hand-programmed to link out to their relative features

Creates new levels of interactivity, allowing users to view the product details using swipe motion gesture

Revolutionize, how companies market, advertise, and sell their products

5/23/2016 7Computer Science & Engg

Page 8: Video interaction through finger tips

Contd...

In a nutshell, the future of digital marketing

5/23/2016 8Computer Science & Engg

Page 9: Video interaction through finger tips

PROPOSED SYSTEM

Consist of following modules :

5/23/2016 9Computer Science & Engg

Swipe Motion Gesture Technique

Video Post-processing

Video Pre-processing

Page 10: Video interaction through finger tips

Contd...

Video Pre –processing Module

Tracking – Learning – Detection (TLD) : A framework designed for long-term tracking of an unknown object in a video stream

ROI is defined by a bounding box in a single frame

TLD simultaneously tracks the object, learns its appearance and detects it whenever it appears in the video

5/23/2016 10Computer Science & Engg

Page 11: Video interaction through finger tips

Contd...

5/23/2016 11Computer Science & Engg

Page 12: Video interaction through finger tips

Contd...

Choosing the video to be pre-processed

5/23/2016 12Computer Science & Engg

Page 13: Video interaction through finger tips

Contd...

Input : Frames of the chosen video

5/23/2016 13Computer Science & Engg

Page 14: Video interaction through finger tips

Contd...

Apply TLD to track interesting objects ( dress, shoes, etc... ) from it

5/23/2016 14Computer Science & Engg

Page 15: Video interaction through finger tips

Contd...

Output : Provides us the locations of the selected objects in each frame

5/23/2016 15Computer Science & Engg

Page 16: Video interaction through finger tips

Contd...

Video Post –processing Module

Input : Output of the TLD algorithm which contains the tracking information

Use that position information to focus the object of interest in each frame

Before the viewer, these frames appear in a rate more than our perception of vision, gives a sensation of moving of the object

5/23/2016 16Computer Science & Engg

Page 17: Video interaction through finger tips

Product Selection Technique

Difficulty of selecting moving objects

5/23/2016 17Computer Science & Engg

Page 18: Video interaction through finger tips

Swipe Motion Gesture

For selecting product, we want to use simple swipe of the color card in front of the webcam

5/23/2016 18Computer Science & Engg

Page 19: Video interaction through finger tips

Contd...

Implementation of swipe motion gesture technique

Take each frame of the video

Smooth the frame using Gaussian blur

Convert frame from BGR to HSV color-space

Extract the specified colored object alone using HSV control panel

Threshold the HSV image for a range of chosen color

5/23/2016 19Computer Science & Engg

Page 20: Video interaction through finger tips

Contd...

Track the colored object and calculate the co-ordinates of the position of the center of the blobs as follows

𝑚𝑝𝑞 = 𝑥=0𝑀−1 𝑦=0

𝑁−1 𝑥𝑝𝑦𝑞𝑓 𝑥, 𝑦

𝑥 = 𝑚10 𝑚00

𝑦 = 𝑚01 𝑚00

5/23/2016 20Computer Science & Engg

Page 21: Video interaction through finger tips

Contd...

Find the distance traveled by the object in first 15 frames

𝑑 = 𝑥2 − 𝑥12 + 𝑦2 − 𝑦1

2

Find velocity of the object

𝑣 = 𝑑 𝑡

If velocity > threshold and the gesture made is a swipe gesture, then the corresponding action is triggered

5/23/2016 21Computer Science & Engg

Page 22: Video interaction through finger tips

Contd...

Identification of blue color card

5/23/2016 22Computer Science & Engg

Page 23: Video interaction through finger tips

RESULT

Output of the proposed system

5/23/2016 23Computer Science & Engg

Page 24: Video interaction through finger tips

CONCLUSION

Uses computer vision technology, TLD algorithm to track unknown objects throughout the video

Employed swipe motion gesture for real time interaction experience

Solution to “How can we increase sales and engagement through traditional video mediums?”

5/23/2016 24Computer Science & Engg

Page 25: Video interaction through finger tips

REFERENCES

[1] Khalad Hasan, Yang Wang, Wing Kwong and Pourang Irani. Enabling User Interactions with Video Contents, 2013[2] Z. Kalal, K. Mikolajczyk, J. Matas, J. Tracking-Learning- Detection. IEEE Transactions on PAMI, vol.34, no.7, pp.1409-1422, July 2012[3] Z. Kalal, J. Matas, and K. Mikolajczyk. P-N Learning: Boot-strapping Binary Classifiers by Structural Constraints. IEEE Confer. on Computer Vision and Pattern Recognition, 2010.[4] Z. Kalal, K. Mikolajczyk, and J. Matas. Forward-Backward Error: Automatic Detection of Tracking Failures, Proc. IAPR International Conference on Pattern Recognition, pp. 23-26,2010.[5] Saikat Basak and Arundhuti Chowdhury . A Vision Interface Framework for Intuitive Gesture Recognition using Color based Blob Detection. International Journal of Computer Applications, vol.90, no.15, March 2014

5/23/2016 25Computer Science & Engg

Page 26: Video interaction through finger tips

ACHIEVEMENTS

Won Best Paper Award for the paper entitled“Video Interaction Through Finger Tips” in theIEEE International Conference on Circuit, Powerand Computing Technologies (ICCPCT-2015),Kanyakumari, Tamilnadu, on March, 2015

Selected to present paper entitled “Turning Existing Video Contents IntoTransactional Content Marketing” at IEEE and SPRINGER Technically Co-Sponsored Science and Information Conference 2015, LONDON , UK

5/23/2016 26Computer Science & Engg

Page 27: Video interaction through finger tips

5/23/2016 27Computer Science & Engg

Thank you for your attention !Questions ?