GOT-10k: A Large High-Diversity Benchmark for Generic Object Tracking in the Wild
By Lianghua Huang, Xin Zhao, Kaiqi Huang. 2018
1,831 citations recorded by the source corpus.
We introduce here a large tracking database that offers an unprecedentedly wide coverage of common moving objects in the wild, called GOT-10k. Specifically, GOT-10k is built upon the backbone of WordNet structure <xref ref-type="bibr" rid="ref1">[1]</xref> and it populates the majority of over 560 classes of moving objects and 87 motion patterns, magnitudes wider than the most recent similar-scale counterparts <xref ref-type="bibr" rid="ref19">[19]</xref>, <xref ref-type="bibr" rid="ref20">[20]</xref>, <xref ref-type="bibr" rid="ref23">[23]</xref>, <xref ref-type="bibr" rid="ref26">[26]</xref>. By releasing the large high-diversity database, we aim to provide a unified training and evaluation platform for the development of class-agnostic, generic purposed short-term trackers. The features of GOT-10k and the contributions of this article are summarized in the following. (1) GOT-10k offers over 10,000 video segments with more than 1.5 million manually labeled bounding boxes, enabling unified training and stable evaluation of deep trackers. (2) GOT-10k is by far the first video trajectory dataset that uses the semantic hierarchy of WordNet to guide class population, which ensures a comprehensive and relatively unbiased coverage of diverse moving objects. (3) For the first time, GOT-10k introduces the <italic>one-shot</italic> protocol for tracker evaluation, where the training and test classes are <italic>zero-overlapped</italic>. The protocol avoids biased evaluation results towards familiar objects and it promotes generalization in tracker development. (4) GOT-10k offers additional labels such as motion classes and object visible ratios, facilitating the development of motion-aware and occlusion-aware trackers. (5) We conduct extensive tracking experiments with 39 typical tracking algorithms and their variants on GOT-10k and analyze their results in this paper. (6) Finally, we develop a comprehensive platform for the tracking community that offers full-featured evaluation toolkits, an online evaluation server, and a responsive leaderboard. The annotations of GOT-10k’s test data are kept private to avoid tuning parameters on it.
Subfields: Computer Vision and Pattern Recognition, Safety, Risk, Reliability and Quality, Evaluation & Benchmarks
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
- Luc Van Gool
- Ming-Hsuan Yang
- Dacheng Tao
- Jitendra Malik
- Xiaogang Wang
- Trevor Darrell
- Andrew Zisserman
- Kristen Grauman
In Safety, Risk, Reliability and Quality
- Martin Danelljan
- Houwen Peng
- Zhipeng Zhang
- Fahad Shahbaz Khan
- Goutam Bhat
- Ming-Hsuan Yang
- Luc Van Gool
- Haibin Ling
In Evaluation & Benchmarks
- Kristen Grauman
- Jitendra Malik
- Abhinav Kumar Gupta
- Luc Van Gool
- Antonio Torralba
- Martin Danelljan
- Ming-Hsuan Yang
- Giovanni Maria Farinella
Citations in the corpus
- Ego4D: Around the World in 3,000 Hours of Egocentric Video (2021, background): “17 [99] Lianghua Huang, Xin Zhao, and Kaiqi Huang.”
- Probabilistic Regression for Visual Tracking (2020, methodology): “GOT10k [23]: This dataset contains 10 , 000 sequences for training and 180 for testing.”
- Deep Learning for Visual Tracking: A Comprehensive Survey (2019, background): “[48], OTB2015 [186], TC128 [187], UAV123 [188], NUS-PRO [189], NfS [190], DTB [191], TrackingNet [192], OxUvA [193], BUAA-PRO [194], GOT10k [195], and LaSOT [196] are com-”
- Is First Person Vision Challenging for Object Tracking? (2020, positive): “Following the increased development of deep learning based trackers, large-scale generic-domain tracking datasets have been recently released [61, 38, 28].”
- Siamese Box Adaptive Network for Visual Tracking (2020, methodology): “• We design a Siamese box adaptive network, which can perform end-to-end offline training with deep convolutional neural networks [12] on well-annotated datasets [34, 30, 25, 15, 9].”
- Perception Test: A Diagnostic Benchmark for Multimodal Video Models (2023, background): “It is defined as the average IoU over the entire track between the predicted and the ground-truth boxes [32, 10].”
- LaSOT: A High-Quality Benchmark for Large-Scale Single Object Tracking (2018, background): “GOT-10k [22] consists of 10,000 videos, aiming to provide rich motion trajectories for developing and evaluating trackers.”
- Transformer Meets Tracker: Exploiting Temporal Context for Robust Visual Tracking (2021, methodology): “Comparison results on the GOT-10k test set [23] in terms of average overlap (AO), and success rates (SR) at overlap thresholds 0.”
- Joint Feature Learning and Relation Modeling for Tracking: A One-Stream Framework (2022, methodology): “The training splits of COCO [30], LaSOT [14], GOT-10k [20] (1k forbidden sequences from GOT-10k training set are removed following the convention [47]) and TrackingNet [35] are used for training.”
- Backbone is All Your Need: A Simplified Architecture for Visual Object Tracking (2022, methodology): “We compare our SimTrack with other trackers on five datasets, including LaSOT [12], TNL2K [48], TrackingNet [35], UAV123 [34] and GOT-10k [22].”
- Stream-51: Streaming Classification and Novelty Detection from Videos (2020, background): “Stream-51 was curated from a variety of existing, videobased object detection and tracking datasets including: the Generic Object Tracking (GOT-10K) dataset [19], the VID dataset [38], and the Large Single Object Tracking (LaSOT) dataset [11].”
- AiATrack: Attention in Attention for Transformer Visual Tracking (2022, methodology): “Tracker Source LaSOT [17] TrackingNet [45] GOT-10k [25] AUC PNorm P AUC PNorm P AO SR0.”
- Graph Attention Tracking (2020, neutral): “We evaluate our SiamGAT on several challenge benchmarks, including GOT-10k [14], OTB-100 [31], UAV123 [21] and LaSOT [7].”
- SiamCAR: Siamese Fully Convolutional Classification and Regression for Visual Tracking (2019, methodology): “The evaluation on GOT-10K [18].”