Self-Supervised Monocular Depth Estimation in Complex Dynamic Environments / (Record no. 617317)
[ view plain ]
| 000 -LEADER | |
|---|---|
| fixed length control field | 01744nam a22001577a 4500 |
| 082 ## - DEWEY DECIMAL CLASSIFICATION NUMBER | |
| Classification number | 629.8 |
| 100 ## - MAIN ENTRY--PERSONAL NAME | |
| Personal name | Ubaid, Bilal |
| 245 ## - TITLE STATEMENT | |
| Title | Self-Supervised Monocular Depth Estimation in Complex Dynamic Environments / |
| 264 ## - PRODUCTION, PUBLICATION, DISTRIBUTION, MANUFACTURE, AND COPYRIGHT NOTICE | |
| Name of producer, publisher, distributor, manufacturer | SMME- NUST; |
| Date of production, publication, distribution, manufacture, or copyright notice | 2026. |
| 300 ## - PHYSICAL DESCRIPTION | |
| Extent | 59p. ; |
| Dimensions | 30cm, |
| 500 ## - GENERAL NOTE | |
| General note | Monocular RGB image-based depth estimation plays an important role for autonomous <br/>driving, 3D reconstruction, robotics, and augmented reality/virtual reality. Self-supervised <br/>monocular depth estimation methods have recently performed impressively in scenes <br/>containing static objects primarily based on the assumption that scenes are consistent when <br/>viewed from different frames. Violations of this occur with moving objects and occlusions, <br/>leading to poor performance in depth accuracy in dynamic scenes and blurry object <br/>boundaries due to exclusion of dynamic areas from the training data. To mitigate such <br/>issues, we offer a self-supervised monocular depth estimation network using channelattention module that incorporates external pre-trained depth estimation models (pseudodepth) into its loss functions and a guided channel-attention mechanism in the decoder of <br/>the depth estimation network. These additions enabled our model to accurately estimate <br/>dynamic objects’ depth with clear boundaries when trained on highly dynamic video <br/>scenes. We tested this approach on the BONN, KITTI, and NYUv2 datasets which contain <br/>both static and highly dynamic scenes. Results indicate that our approach performs <br/>competitively with prior approaches. <br/> |
| 650 ## - SUBJECT ADDED ENTRY--TOPICAL TERM | |
| Topical term or geographic name entry element | MS R & AI |
| 700 ## - ADDED ENTRY--PERSONAL NAME | |
| Personal name | Supervisor: Dr. Shahbaz Khan |
| 856 ## - ELECTRONIC LOCATION AND ACCESS | |
| Uniform Resource Identifier | <a href=" http://10.250.8.41:8080/xmlui/handle/123456789/57587"> http://10.250.8.41:8080/xmlui/handle/123456789/57587</a> |
| 942 ## - ADDED ENTRY ELEMENTS (KOHA) | |
| Source of classification or shelving scheme | |
| Koha item type | Soft Copy |
| Withdrawn status | Permanent Location | Current Location | Shelving location | Date acquired | Full call number | Barcode | Koha item type |
|---|---|---|---|---|---|---|---|
| School of Mechanical & Manufacturing Engineering (SMME) | School of Mechanical & Manufacturing Engineering (SMME) | Thesis | 05/06/2026 | 629.8 | SMME-TH-1214 | Soft Copy |