Dimension Decoupled Region Proposal Network For Object Detection

2023-1
Baytekin, Mehmet Can
Show full item record
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
141
views
74
downloads
In this work, we propose a dynamic anchor, dynamic assignment region proposal network named DA2RPN to improve the traditional Region Proposal Network (RPN). Classical Region Proposal Network places a set of pre-determined anchor boxes to generate object proposals which are later consumed by detection heads to produce final results. Problems of the Region Proposal Network are for each feature map point same anchor boxes are placed and static threshold values based on intersection-overunion scores are used to label anchor boxes. However, anchors mostly residing out of the image plane are not that useful to detect objects as objects only reside in the image plane. On the other hand, anchor box-ground truth box pairs having the same intersection-over-union values may not be equally useful for the detection of different objects. One anchor can be the best candidate for that object but another anchor can be a poor choice for another object even if they have the same iou values. To mitigate these problems, we generate a different number of anchors per feature map point and use a dynamic thresholding mechanism adaptive to the quality of selected aniv chors per ground-truth box. Additionally, to ease the training and to prevent sparsity caused by dynamic anchor generation we decouple anchor dimensions. Our results experimented on the COCO dataset show improvements over the Region Proposal Network
object detection, region proposal network, dynamic anchor box, dynamic assignment
https://hdl.handle.net/11511/102075
Graduate School of Informatics, Thesis

Suggestions

Density-Aware Probabilistic Clustering in Ad Hoc Networks
Ergenç, Doǧanalp; Eksert, Levent; Onur, Ertan (2018-06-07)
Clustering makes an ad hoc network scalable forming easy-to-manage local groups. However, it brings an extra control overhead to create and maintain clustered network topology. In this paper, we propose Probabilistic Clustering Algorithm that is a simple and efficient clustering algorithm with minimal overhead. In this algorithm, cluster heads are determined probabilistically in a distributed fashion. An analytic model is introduced for nodes to compute the probability of declaring themselves as cluster hea...
DISTRIBUTED 3D DYNAMIC MESH CODING
Bici, M. Oguz; Akar, Gözde (2008-10-15)
In this paper, we propose a distributed 3D dynamic mesh coding system. The system is based on Slepian and Wolf's and Wyner and Ziv's information-theoretic results. Our system extends the ideas in distributed video coding to 3D dynamic meshes with constant connectivity. The connectivity of the sequence and key frames are encoded and decoded by a conventional static mesh coder. The Wyner-Ziv frames are encoded independent of key frames but decoded jointly with decoded key frames. The joint decoding is perform...
Artificial intelligence based dynamic mission planning with probabilistic roadmaps and Voronoi diagrams using predictive launch acceptability region approach
Özdemir, Mustafa Raşit; Ertekin Bolelli, Şeyda; Department of Computer Engineering (2021-9-2)
In this thesis, dynamic air-to-surface mission planning strategies based on probabilistic roadmaps and Voronoi diagrams using predictive launch acceptability region approach are proposed for opportunity targets in order to strengthen decision support capabilities of aircraft. Air-to-surface missions are planned in ground support systems and loaded to aircraft before the mission begins. This means that all the waypoints which should be followed during an air-to-surface mission are planned according to va...
Compact and Simple RLWE Based Key Encapsulation Mechanism
Alkim, Erdem; Bilgin, Yusuf Alper; Cenk, Murat (2019-01-01)
n this paper, we propose a key encapsulation scheme based on NewHope and Kyber, two NIST post-quantum standardization project candidates. Our scheme is based on NewHope, thus it is simple and has fast implementation while it is making use of smaller key sizes and easily changeable security level advantages of Kyber. The scheme heavily use recent advances on Number Theoretic Transform (NTT) in a way that transformation from one degree polynomial to another is easy. To make it possible, we changed the definit...
Reachability analysis using proximity based automata
Kapinski, Jim; Schmidt, Klaus Verner; Krogh, Bruce (2004-01-01)
This paper introduces proximity based automata (PBA) to perform reachability analysis for discrete-time, continuous-state systems. The notion of proximity of states in the continuous valued state space is used to evaluate the response of the system efficiently. A PBA is constructed using ellipsoidal techniques and tools for solving optimization problems constrained by linear matrix inequalities. Several empirical studies demonstrate how trade-offs between conservativeness and computational complexity can be...
Citation Formats
M. C. Baytekin, “Dimension Decoupled Region Proposal Network For Object Detection,” M.S. - Master of Science, Middle East Technical University, 2023.
METU IIS - Integrated Information Service open@metu.edu.tr