Action-Decision Networks for Visual Tracking with Deep Reinforcement Learning

By Sangdoo Yun, Jongwon Choi, Y. Yoo, Kimin Yun, Jin Young Choi. Computer Vision and Pattern Recognition, 2017

509 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Global and Planetary Change

View Paper: https://www.semanticscholar.org/paper/96dc41d3b004fd4c7f96b71b4e174beb3088b2bb

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Global and Planetary Change

Citations in the corpus

Part-Activated Deep Reinforcement Learning for Action Prediction (2018, background): “Deep Reinforcement Learning: Recently, the field of reinforcement learning resurrects with the strong support from deep learning [13, 18, 45].”
Learning Aberrance Repressed Correlation Filters for Real-Time UAV Tracking (2019, methodology): “The results are compared with 20 state-of-the-art trackers with both hand-crafted based trackers and deep-based trackers, i.e., KCF [13], DSST [8], SAMF [18], MUSTER [14], BACF [15], SRDCF [7], STAPLE CA [21], MCCT-H [27], STRCF [16], ECO-HC (with gray-scale) [5], ECO [5], CCOT [9], HCF [19], ADNet ”
VITAL: VIsual Tracking via Adversarial Learning (2018, unknown): “…MEEM [63], RPT [34], LCT [37], MUSTer [25], HCFT [36], FCNT [57], SRDCF [12], CNN-SVM [24], DeepSRDCF [11], DAT [41], Staple [3], SRDCFdecon [13], CCOT [14], GOTURN [21], SINT [54], SiamFC [4], HDT [43], SCT [6], MDNet [38], DLS-SVM [39], ADNet [61], ECO [9], MCPF [64], CFNet [55] and CREST [50].”
Real-Time MDNet (2018, methodology): “Recently, Convolutional Neural Networks (CNNs) are very effective in visual tracking [1,2,3,4,5,6,7,8,9,10], but, unfortunately, highly accurate tracking algorithms based on CNNs are often too slow for practical systems.”
Reinforced Attention for Few-Shot Learning and Beyond (2021, methodology): “Reinforcement learning is easily utilized in sequential visual tasks, like tracking [64, 3, 41], action recognition [63, 52, 4], visual navigation [57, 67], video summarization [29] and person re-id [25].”
AutoTrack: Towards High-Performance Visual Tracking for UAV With Automatic Spatio-Temporal Regularization (2020, methodology): “Compared to deep trackers including ASRCF [31], TADT [51], SiameseFC [15], DSiam [56], MCCT [13], ADNet [16], CFNet [52], DeepSTRCF [14], UDT+ [53], HCF [18], C-COT [17], ECO [20], IBCCF [54], MCPF [19] and CREST [35], AutoTrack with a single CPU exhibits the”
GOT-10k: A Large High-Diversity Benchmark for Generic Object Tracking in the Wild (2018, background): “s, the tracking community has also explored other advanced deep models. For example, some works view tracking as a sequential decision making task and learn policies to guide the online actions [59], [66]. Some other works explore recurrent structures for sequential prediction in tracking [64], [65]”
Attentional Correlation Filter Network for Adaptive Visual Tracking (2017, background): “ers [29, 33, 35, 36, 41] which require frequent fine-tuning of the network to learn the appearance of the target.”
From Discriminant to Complete: Reinforcement Searching-Agent Learning for Weakly Supervised Object Detection (2020, background): “Recently, several successful attempts have been made to apply the DRL methods in the computer vision community [14]–[18], [32].”
Is First Person Vision Challenging for Object Tracking? (2020, unknown): “This kind of representation has been used in deep regression networks [36, 32], online tracking-by-detection methods [62, 73], approaches based on reinforcement learning [88, 14, 26, 33], deep discriminative correlation filters [21, 22, 9, 24, 54, 10], and trackers based on siamese networks [8, 48, ”
MASK-RL: Multiagent Video Object Segmentation Framework Through Reinforcement Learning (2020, methodology): “For these reasons, RL has recently attracted a lot of attention for emulating human behavior in several contexts, such as games [21], [29] and autonomous driving [22], but also for computer vision tasks [1], [13], [19], [25], [27], [47], including video object segmentation [11], [12], [37].”
Towards More Flexible and Accurate Object Tracking with Natural Language: Algorithms and Benchmark (2021, unknown): “300 034 ADNet [60] CVPR-2017 CNN-3 3 3 Local 3@GPU 0.”
Deep Learning for Visual Tracking: A Comprehensive Survey (2019, methodology): “[162], [172]–[174], [174], [175], [177], [181].”
Target-Aware Deep Tracking (2019, background): “On the other hand, numerous trackers [16, 6, 28, 17, 35, 47, 12] are developed by improving the decision models including support vector machines, correlation filters, deep classifiers, and deep regressors.”
Deep Progressive Reinforcement Learning for Skeleton-Based Action Recognition (2018, methodology): “[36] employed DRL for visual tracking and Rao et al.”