| 000 -LEADER |
|---|
| fixed length control field |
02730nam a22001577a 4500 |
| 082 ## - DEWEY DECIMAL CLASSIFICATION NUMBER |
|---|
| Classification number |
629.8 |
| 100 ## - MAIN ENTRY--PERSONAL NAME |
|---|
| Personal name |
ALI, HASNAIN |
| 245 ## - TITLE STATEMENT |
|---|
| Title |
Multimedia Analytics for Scene Content Understanding / |
| Statement of responsibility, etc. |
HASNAIN ALI |
| 264 ## - PRODUCTION, PUBLICATION, DISTRIBUTION, MANUFACTURE, AND COPYRIGHT NOTICE |
|---|
| Place of production, publication, distribution, manufacture |
Islamabad : |
| Name of producer, publisher, distributor, manufacturer |
SMME- NUST; |
| Date of production, publication, distribution, manufacture, or copyright notice |
2025. |
| 300 ## - PHYSICAL DESCRIPTION |
|---|
| Extent |
133p. |
| Other physical details |
Soft Copy |
| Dimensions |
30cm |
| 500 ## - GENERAL NOTE |
|---|
| General note |
With the rapid expansion of video content, understanding how humans retain and recall visual data has become crucial. Memorability, a key neurocognitive process, plays<br/>a significant role in retaining and retrieving video content. While past research has<br/>explored image memorability, video memorability has received less attention, leaving a<br/>gap in robust computational models for predicting memorable video events. This thesis<br/>addresses this gap through a multi-phase study focused on video memorability prediction, scalable feature extraction, and behavior training for robotic systems. The first<br/>study introduces a novel framework that predicts episodic video memorability by fusing deep features, including text, color, and motion. Episodic sequences are generated<br/>using a Fuzzy FastText model and color histogram analysis, while scene objects are<br/>identified using a Faster Region-based Convolutional Neural Network (Faster R-CNN).<br/>The fusion of these features results in improved short- and long-term memorability, with<br/>a superior Spearman’s rank correlation of 0.6428 and 0.4285, respectively. The second<br/>study focuses on a robust Stacked Bin-Convolutional Neural Network (SB-CNN) and<br/>Sparse Low-Rank Regressor (SLRR) model. This model improves video event classification by employing a low-rank representation technique that reduces noise in video<br/>frames, leading to more accurate predictions. The Multi-Attribute Decision Making<br/>(MADM) technique is applied to enhance decision making, achieving a recall time of<br/>49.9247 on public datasets. In the final study, a Trimmed Q-learning algorithm is introduced to optimize memorability-driven scene prediction in mobile robots. The training<br/>is conducted through online, short-term, and long-term learning modules, with significant improvements in memorability scores: 72.84% for short-term and online learning,<br/>and 68.63% for long-term learning. By linking these phases, this thesis presents an<br/>integrated framework that effectively addresses the challenges of video memorability<br/>prediction, robust feature scaling, and robotic decision-making, offering practical insights for both academic research and real-world applications. |
| 650 ## - SUBJECT ADDED ENTRY--TOPICAL TERM |
|---|
| Topical term or geographic name entry element |
PhD Robotics and Intelligent Machine Engineering |
| 700 ## - ADDED ENTRY--PERSONAL NAME |
|---|
| Personal name |
Supervisor : Dr Syed Omer Gilani |
| 856 ## - ELECTRONIC LOCATION AND ACCESS |
|---|
| Uniform Resource Identifier |
<a href="http://10.250.8.41:8080/xmlui/handle/123456789/56235">http://10.250.8.41:8080/xmlui/handle/123456789/56235</a> |
| 942 ## - ADDED ENTRY ELEMENTS (KOHA) |
|---|
| Source of classification or shelving scheme |
|
| Koha item type |
Thesis |
