2022
Gupta, Akshita; Narayan, Sanath; Joseph, K J; Khan, Salman; Khan, Fahad Shahbaz; Shah, Mubarak
OW-DETR: Open-world Detection Transformer Conference
IEEE Computer Vision and Pattern Recognition, 2022.
Abstract | Tags: CVPR, Object Detection, Open World, Transformers | Links:
@conference{nokey_109,
title = {OW-DETR: Open-world Detection Transformer},
author = {Akshita Gupta and Sanath Narayan and K J Joseph and Salman Khan and Fahad Shahbaz Khan and Mubarak Shah},
url = {https://www.crcv.ucf.edu/wp-content/uploads/2018/11/03815.pdf
https://www.crcv.ucf.edu/wp-content/uploads/2018/11/03815-supp.pdf
https://github.com/akshitac8/OW-DETR.},
year = {2022},
date = {2022-06-19},
urldate = {2022-06-19},
publisher = {IEEE Computer Vision and Pattern Recognition},
abstract = {Open-world object detection (OWOD) is a challenging computer vision problem, where the task is to detect a known set of object categories while simultaneously identifying unknown objects. Additionally, the model must incrementally learn new classes that become known in the next training episodes. Distinct from standard object detection, the OWOD setting poses significant challenges for generating quality candidate proposals on potentially unknown objects, separating the unknown objects from the background and detecting diverse unknown objects. Here, we introduce a novel end-to-end transformer-based framework, OW-DETR, for open-world object detection. The proposed OW-DETR comprises three dedicated components
namely, attention-driven pseudo-labeling, novelty classification and objectness scoring to explicitly address the aforementioned OWOD challenges. Our OW-DETR explicitly encodes multi-scale contextual information, possesses less inductive bias, enables knowledge transfer from known classes to the unknown class and can better discriminate between unknown objects and background. Comprehensive experiments are performed on two benchmarks: MS-COCO and PASCAL VOC. The extensive ablations reveal the merits of our proposed contributions. Further, our model outperforms the recently introduced OWOD approach, ORE, with absolute gains ranging from 1.8% to 3.3% in terms of unknown recall on MS-COCO. In the case of incremental
object detection, OW-DETR outperforms the state-of-the art for all settings on PASCAL VOC. Our code is available at https://github.com/akshitac8/OW-DETR.},
keywords = {CVPR, Object Detection, Open World, Transformers},
pubstate = {published},
tppubtype = {conference}
}
Open-world object detection (OWOD) is a challenging computer vision problem, where the task is to detect a known set of object categories while simultaneously identifying unknown objects. Additionally, the model must incrementally learn new classes that become known in the next training episodes. Distinct from standard object detection, the OWOD setting poses significant challenges for generating quality candidate proposals on potentially unknown objects, separating the unknown objects from the background and detecting diverse unknown objects. Here, we introduce a novel end-to-end transformer-based framework, OW-DETR, for open-world object detection. The proposed OW-DETR comprises three dedicated components
namely, attention-driven pseudo-labeling, novelty classification and objectness scoring to explicitly address the aforementioned OWOD challenges. Our OW-DETR explicitly encodes multi-scale contextual information, possesses less inductive bias, enables knowledge transfer from known classes to the unknown class and can better discriminate between unknown objects and background. Comprehensive experiments are performed on two benchmarks: MS-COCO and PASCAL VOC. The extensive ablations reveal the merits of our proposed contributions. Further, our model outperforms the recently introduced OWOD approach, ORE, with absolute gains ranging from 1.8% to 3.3% in terms of unknown recall on MS-COCO. In the case of incremental
object detection, OW-DETR outperforms the state-of-the art for all settings on PASCAL VOC. Our code is available at https://github.com/akshitac8/OW-DETR.
namely, attention-driven pseudo-labeling, novelty classification and objectness scoring to explicitly address the aforementioned OWOD challenges. Our OW-DETR explicitly encodes multi-scale contextual information, possesses less inductive bias, enables knowledge transfer from known classes to the unknown class and can better discriminate between unknown objects and background. Comprehensive experiments are performed on two benchmarks: MS-COCO and PASCAL VOC. The extensive ablations reveal the merits of our proposed contributions. Further, our model outperforms the recently introduced OWOD approach, ORE, with absolute gains ranging from 1.8% to 3.3% in terms of unknown recall on MS-COCO. In the case of incremental
object detection, OW-DETR outperforms the state-of-the art for all settings on PASCAL VOC. Our code is available at https://github.com/akshitac8/OW-DETR.
Gupta, Akshita; Narayan, Sanath; Joseph, K J; Khan, Salman; Khan, Fahad Shahbaz; Shah, Mubarak
OW-DETR: Open-world Detection Transformer Conference
IEEE Computer Vision and Pattern Recognition, 2022.
Abstract | Tags: CVPR, Object Detection, Open World, Transformers | Links:
@conference{nokey_157,
title = {OW-DETR: Open-world Detection Transformer},
author = {Akshita Gupta and Sanath Narayan and K J Joseph and Salman Khan and Fahad Shahbaz Khan and Mubarak Shah},
url = {https://www.crcv.ucf.edu/wp-content/uploads/2018/11/03815.pdf
https://www.crcv.ucf.edu/wp-content/uploads/2018/11/03815-supp.pdf
https://github.com/akshitac8/OW-DETR.},
year = {2022},
date = {2022-06-19},
urldate = {2022-06-19},
publisher = {IEEE Computer Vision and Pattern Recognition},
abstract = {Open-world object detection (OWOD) is a challenging computer vision problem, where the task is to detect a known set of object categories while simultaneously identifying unknown objects. Additionally, the model must incrementally learn new classes that become known in the next training episodes. Distinct from standard object detection, the OWOD setting poses significant challenges for generating quality candidate proposals on potentially unknown objects, separating the unknown objects from the background and detecting diverse unknown objects. Here, we introduce a novel end-to-end transformer-based framework, OW-DETR, for open-world object detection. The proposed OW-DETR comprises three dedicated components
namely, attention-driven pseudo-labeling, novelty classification and objectness scoring to explicitly address the aforementioned OWOD challenges. Our OW-DETR explicitly encodes multi-scale contextual information, possesses less inductive bias, enables knowledge transfer from known classes to the unknown class and can better discriminate between unknown objects and background. Comprehensive experiments are performed on two benchmarks: MS-COCO and PASCAL VOC. The extensive ablations reveal the merits of our proposed contributions. Further, our model outperforms the recently introduced OWOD approach, ORE, with absolute gains ranging from 1.8% to 3.3% in terms of unknown recall on MS-COCO. In the case of incremental
object detection, OW-DETR outperforms the state-of-the art for all settings on PASCAL VOC. Our code is available at https://github.com/akshitac8/OW-DETR.},
keywords = {CVPR, Object Detection, Open World, Transformers},
pubstate = {published},
tppubtype = {conference}
}
Open-world object detection (OWOD) is a challenging computer vision problem, where the task is to detect a known set of object categories while simultaneously identifying unknown objects. Additionally, the model must incrementally learn new classes that become known in the next training episodes. Distinct from standard object detection, the OWOD setting poses significant challenges for generating quality candidate proposals on potentially unknown objects, separating the unknown objects from the background and detecting diverse unknown objects. Here, we introduce a novel end-to-end transformer-based framework, OW-DETR, for open-world object detection. The proposed OW-DETR comprises three dedicated components
namely, attention-driven pseudo-labeling, novelty classification and objectness scoring to explicitly address the aforementioned OWOD challenges. Our OW-DETR explicitly encodes multi-scale contextual information, possesses less inductive bias, enables knowledge transfer from known classes to the unknown class and can better discriminate between unknown objects and background. Comprehensive experiments are performed on two benchmarks: MS-COCO and PASCAL VOC. The extensive ablations reveal the merits of our proposed contributions. Further, our model outperforms the recently introduced OWOD approach, ORE, with absolute gains ranging from 1.8% to 3.3% in terms of unknown recall on MS-COCO. In the case of incremental
object detection, OW-DETR outperforms the state-of-the art for all settings on PASCAL VOC. Our code is available at https://github.com/akshitac8/OW-DETR.
namely, attention-driven pseudo-labeling, novelty classification and objectness scoring to explicitly address the aforementioned OWOD challenges. Our OW-DETR explicitly encodes multi-scale contextual information, possesses less inductive bias, enables knowledge transfer from known classes to the unknown class and can better discriminate between unknown objects and background. Comprehensive experiments are performed on two benchmarks: MS-COCO and PASCAL VOC. The extensive ablations reveal the merits of our proposed contributions. Further, our model outperforms the recently introduced OWOD approach, ORE, with absolute gains ranging from 1.8% to 3.3% in terms of unknown recall on MS-COCO. In the case of incremental
object detection, OW-DETR outperforms the state-of-the art for all settings on PASCAL VOC. Our code is available at https://github.com/akshitac8/OW-DETR.
2021
Ashraf, Muhammad Waseem; Sultani, Waqas; Shah, Mubarak
Dogfight: Detecting Drones from Drones Videos Conference
IEEE Conference on Computer Vision and Pattern Recognition, 2021.
Tags: CVPR, Drone Video Analysis, Object Detection, UAV Video Analysis, UVA Video Analysis | Links:
@conference{Sultani2021,
title = {Dogfight: Detecting Drones from Drones Videos},
author = {Muhammad Waseem Ashraf and Waqas Sultani and Mubarak Shah},
url = {https://www.crcv.ucf.edu/wp-content/uploads/2018/11/Dogfight.pdf
https://www.crcv.ucf.edu/wp-content/uploads/2018/11/Dogfight_Supp.mp4
https://github.com/mwaseema/Drone-Detection
http://im.itu.edu.pk/dogfight-detecting-drones-from-drones-videos/ https://docs.google.com/presentation/d/1huBSbYzyNUCs-gJHdSU2CZ0XfRlBWGOQ/edit#slide=id.p1},
year = {2021},
date = {2021-06-19},
urldate = {2021-06-19},
publisher = {IEEE Conference on Computer Vision and Pattern Recognition},
keywords = {CVPR, Drone Video Analysis, Object Detection, UAV Video Analysis, UVA Video Analysis},
pubstate = {published},
tppubtype = {conference}
}
Ashraf, Muhammad Waseem; Sultani, Waqas; Shah, Mubarak
Dogfight: Detecting Drones from Drones Videos Conference
IEEE Conference on Computer Vision and Pattern Recognition, 2021.
Tags: CVPR, Drone Video Analysis, Object Detection, UAV Video Analysis, UVA Video Analysis | Links:
@conference{Sultani2021b,
title = {Dogfight: Detecting Drones from Drones Videos},
author = {Muhammad Waseem Ashraf and Waqas Sultani and Mubarak Shah},
url = {https://www.crcv.ucf.edu/wp-content/uploads/2018/11/Dogfight.pdf
https://www.crcv.ucf.edu/wp-content/uploads/2018/11/Dogfight_Supp.mp4
https://github.com/mwaseema/Drone-Detection
http://im.itu.edu.pk/dogfight-detecting-drones-from-drones-videos/ https://docs.google.com/presentation/d/1huBSbYzyNUCs-gJHdSU2CZ0XfRlBWGOQ/edit#slide=id.p1},
year = {2021},
date = {2021-06-19},
urldate = {2021-06-19},
publisher = {IEEE Conference on Computer Vision and Pattern Recognition},
keywords = {CVPR, Drone Video Analysis, Object Detection, UAV Video Analysis, UVA Video Analysis},
pubstate = {published},
tppubtype = {conference}
}
2014
Ardeshir, Shervin; Zamir, Amir Roshan; Torroella, Alejandro; Shah, Mubarak
GIS-Assisted Object Detection and Geospatial Localization Conference
ECCV 2014, 2014.
Tags: Deep Learning, Geo-Localization, GIS, Object Detection, REU | Links:
@conference{Ardeshir2015b,
title = {GIS-Assisted Object Detection and Geospatial Localization},
author = {Shervin Ardeshir and Amir Roshan Zamir and Alejandro Torroella and Mubarak Shah},
url = {https://www.crcv.ucf.edu/papers/GIS_ECCV14.pdf},
year = {2014},
date = {2014-01-25},
urldate = {2014-01-25},
publisher = {ECCV 2014},
keywords = {Deep Learning, Geo-Localization, GIS, Object Detection, REU},
pubstate = {published},
tppubtype = {conference}
}
Ardeshir, Shervin; Zamir, Amir Roshan; Torroella, Alejandro; Shah, Mubarak
GIS-Assisted Object Detection and Geospatial Localization Conference
ECCV 2014, 2014.
Tags: Deep Learning, Geo-Localization, GIS, Object Detection, REU | Links:
@conference{Ardeshir2015bb,
title = {GIS-Assisted Object Detection and Geospatial Localization},
author = {Shervin Ardeshir and Amir Roshan Zamir and Alejandro Torroella and Mubarak Shah},
url = {https://www.crcv.ucf.edu/papers/GIS_ECCV14.pdf},
year = {2014},
date = {2014-01-25},
urldate = {2014-01-25},
publisher = {ECCV 2014},
keywords = {Deep Learning, Geo-Localization, GIS, Object Detection, REU},
pubstate = {published},
tppubtype = {conference}
}
2007
Yan, Pingkun; Khan, Saad M.; Shah, Mubarak
3D Model based Object Class Detection in an Arbitrary View Conference
ICCV, 2007.
Tags: Deep Learning, Multi-view, Object Detection, Rendering | Links:
@conference{Yan2007,
title = {3D Model based Object Class Detection in an Arbitrary View},
author = {Pingkun Yan and Saad M. Khan and Mubarak Shah},
url = {https://www.crcv.ucf.edu/papers/PingkunICCV07.pdf},
year = {2007},
date = {2007-11-28},
urldate = {2007-11-28},
publisher = {ICCV},
keywords = {Deep Learning, Multi-view, Object Detection, Rendering},
pubstate = {published},
tppubtype = {conference}
}