A PRELIMINARY REPORT ON

“PREDICTING HEAD MOVEMENT IN PANORAMIC

VIDEO”

SUBMITTED TO THE SAVITRIBAI PHULE PUNE UNIVERSITY, PUNE IN

THE PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE AWARD OF
THE DEGREE

OF

BACHELOR OF ENGINEERING (COMPUTER ENGINEERING)

SUBMITTED BY

KARDILE SAGAR B. 71748917F

KULKARNI UTKARSHA R. 71724114K

MANE VARSHA S. 71628348E

PATNE SHREEYASH A. 71847482B

DEPARTMENT OF COMPUTER ENGINEERING

D.Y. PATIL INSTITUTE OF ENGINEERING & TECHNOLOGY

AMBI, TALEGAON MIDC RD, PUNE, 410506

SAVITRIBAI PHULE PUNE UNIVERSITY 2018 -2019

 CERTIFICATE

This is to certify that the project report entitles

“PREDICTING HEAD MOVEMENT IN PANORAMIC

VIDEO”
Submitted by

KARDILE SAGAR B. 71748917F

KULKARNI UTKARSHA R. 71724114K

MANE VARSHA S. 71628348E

PATNE SHREEYASH A. 71847482B

Are a bonafide student of this institute and the work has been carried out by him/her
under the supervision of Prof. Sharmila Chopade and it is approved for the partial
fulfillment of the requirement of Savitribai Phule Pune University, for the award of the
degree of Bachelor of Engineering (Computer Engineering).

(Prof. Sharmila Chopade) (Prof. Mangesh Manake)

Guide, Head,
Department of Computer Engineering Department of Computer Engineering

External examiner (Dr.L.V. Kamble)

Principal,
D.Y. Patil Institute of Engineering & Technology.Ambi

Place : Pune
Date :
 ACKNOWLEDGEMENT

I respect and thank the guide Prof. Sharmila Chopade, for providing us an opportunity to do the
project work and give us support and guidance which made us complete the report duly. We are
really thankful to her for providing such a nice help in guidance and her support although she
had a busy schedule managing the corporate affairs.

KARDILE SAGAR B.

KULKARNI UTKARSHA R.

MANE VARSHA S.

PATNE SHREEYASH A.
 ABSTRACT

Panoramic video provides immersive and interactive experience by enabling

humans to control the field of view (FoV) through head movement (HM). Thus,
HM plays a key role in modeling human attention on panoramic video.

Here, we establish a database collecting subjects’ HM in panoramic video

sequences. From this database, we find that the HM data are highly consistent
across subjects. Furthermore, we find that deep reinforcement learning (DRL) can
be applied to predict HM positions, via maximizing the reward of imitating human
HM scanpaths through the agent’s actions. Based on our findings, we propose a
DRL-based HM prediction (DHP) approach with offline version called offline-
DHP. In offline-DHP, multiple DRL workflows are run to determine potential HM
positions at each panoramic frame. Then, a heat map of the potential HM positions,
named the HM map, is generated as the output of offline-DHP. Finally, the
experiments validate that our approach is effective in offline prediction of HM
positions for panoramic video, and that the learned offline-DHP model can
improve the performance of online-DHP.
 TABLE OF CONTENTS

Sr.no Contents Pg. No

1. Introduction

1.1 Motivation

1.2 Problem Definition

2. Literature Survey

3. Software Requirement Specification

3.1 Introduction

3.1.1 Project Scope

3.2 Functional requirements

3.3 External Interface requirements

3.3.1 Hardware Interfaces

3.3.2 Software Interfaces

3.3.3 Communication Interfaces

3.4 Non-functional requirements

3.4.1 Performance requirement

3.4.2 Safety requirement

3.4.3 Security requirement

3.4.4 Software Quality Attribute

3.5 System Requirement

3.5.1 Database Requirement

3.5.2 Software requirement

3.5.3 Hardware requirement

 3.6 Analysis Models

3.7 System Implementation Plan

4. System Design

4.1 System Architecture

4.2 Dataflow Diagrams

4.3 Entity Relationship Diagram

4.4 UML Diagrams

5. Other Specifications

5.1 Advantages

5.2 Limitations

5.3 Applications

6. Conclusion and Future Work