Active Object Localization with Deep Reinforcement Learning

Caicedo, Juan C.; Lazebnik, Svetlana

doi:10.1109/iccv.2015.286

Cited by 405 publications

(228 citation statements)

References 41 publications

Supporting

Mentioning

216

Contrasting

Unclassified

Order By: Relevance

“…With deep reinforcement learning, the agent in [7] learned a policy to locate a fixed number of objects through gradually narrowing the bounding box from the hole image to the ground truth. Actions in [7] can change the scale, location or aspect ratio of bounding boxes. The work in [8] reduced the number of actions to six, and it made the policy easier to optimize.…”

Section: Deep Reinforcement Learningmentioning

confidence: 99%

“…The work in [8] reduced the number of actions to six, and it made the policy easier to optimize. Like [7], the detection framework in [8] also used the inhibition-of-return mechanism to locate a fixed number of objects. Different from [7], the agent in [8] adopted the hierarchical representation, which preformed the top-down search to locate objects.…”

Section: Deep Reinforcement Learningmentioning

confidence: 99%

“…Different from the methods with supervised learning, the approaches based on reinforcement learning attract more and more attention. The reinforcement learning proves a new method to handle traditional computer vision problems, such as visual tracking [24], action recognition [25] and object detection [7][8][9], by employing the deep reinforcement learning algorithms.…”

Section: Deep Reinforcement Learningmentioning

confidence: 99%

“…The works in [7][8][9] took the object detection problem as the Markov Decision Process and trained detection agent based on reinforcement learning. The detection agent adopts the top-down searching process that firstly analyses the global image and then narrows down the local regions that contain object information step by step.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

An Aircraft Detection Framework Based on Reinforcement Learning and Convolutional Neural Networks in Remote Sensing Images

Sun

et al. 2018

Remote Sensing

View full text Add to dashboard Cite

Aircraft detection has attracted increasing attention in the field of remote sensing image analysis. Complex background, illumination change and variations of aircraft kind and size in remote sensing images make the task challenging. In our work, we propose an effective aircraft detection framework based on reinforcement learning and a convolutional neural network (CNN) model. Aircraft in remote sensing images can be accurately and robustly located with the help of the searching mechanism that the candidate region is dynamically reduced to the correct location of aircraft, which is implemented through reinforcement learning. The detection framework overcomes the difficulties that the current detection methods based on reinforcement learning are only able to detect a fixed number of objects. Specifically, we adopt the restricted EdgeBoxes that generate the high-quality candidate boxes through the prior aircraft knowledge at first. Then, we train an intelligent detection agent through reinforcement learning and apprenticeship learning. The detection agent accurately locates the aircraft in the candidate boxes within several actions, and it even performs better than the greed strategy in apprenticeship learning. During the final detection step, we carefully design the CNN model that predicts the probability that the localization result generated by the detection agent is an aircraft. Comparative experiments demonstrate the accuracy and efficiency of our aircraft detection framework.

show abstract

Section: Deep Reinforcement Learningmentioning

confidence: 99%

Section: Deep Reinforcement Learningmentioning

confidence: 99%

Section: Deep Reinforcement Learningmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An Aircraft Detection Framework Based on Reinforcement Learning and Convolutional Neural Networks in Remote Sensing Images

Sun

et al. 2018

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Deep fitted q-iteration was proposed for this object. A class-based active detection model was proposed in [4]. It learns to localize object known by the system.…”

Section: Introductionmentioning

confidence: 99%

Hierarchical extreme learning machine based reinforcement learning for goal localization

AlDahoul

Htike

Akmeliawati

2017

IOP Conf. Ser.: Mater. Sci. Eng.

View full text Add to dashboard Cite

Localization and the connection between Uq(so(3)) and Uq(osp(1|2) Abstract. The objective of goal localization is to find the location of goals in noisy environments. Simple actions are performed to move the agent towards the goal. The goal detector should be capable of minimizing the error between the predicted locations and the true ones. Few regions need to be processed by the agent to reduce the computational effort and increase the speed of convergence. In this paper, reinforcement learning (RL) method was utilized to find optimal series of actions to localize the goal region. The visual data, a set of images, is high dimensional unstructured data and needs to be represented efficiently to get a robust detector. Different deep Reinforcement models have already been used to localize a goal but most of them take long time to learn the model. This long learning time results from the weights fine tuning stage that is applied iteratively to find an accurate model. Hierarchical Extreme Learning Machine (H-ELM) was used as a fast deep model that doesn't fine tune the weights. In other words, hidden weights are generated randomly and output weights are calculated analytically. H-ELM algorithm was used in this work to find good features for effective representation. This paper proposes a combination of Hierarchical Extreme learning machine and Reinforcement learning to find an optimal policy directly from visual input. This combination outperforms other methods in terms of accuracy and learning speed. The simulations and results were analysed by using MATLAB.

show abstract

Intrusion Detection Using Data Mining

Astha¹,

Singh²

2021

Artificial Intelligence and Data Mining Approaches in Security Frameworks

View full text Add to dashboard Cite

Active Object Localization with Deep Reinforcement Learning

Cited by 405 publications

References 41 publications

An Aircraft Detection Framework Based on Reinforcement Learning and Convolutional Neural Networks in Remote Sensing Images

An Aircraft Detection Framework Based on Reinforcement Learning and Convolutional Neural Networks in Remote Sensing Images

Hierarchical extreme learning machine based reinforcement learning for goal localization

Intrusion Detection Using Data Mining

Contact Info

Product

Resources

About