The Way They Move: Tracking Multiple Targets with Similar Appearance

By Caglayan Dicle, Octavia I. Camps, Mario Sznaier. IEEE International Conference on Computer Vision, 2013

332 citations recorded by the source corpus.

Subfields: Automotive Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/b238f3f2a487271973c573634611229c432cf467

How researchers see this paper

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

In Automotive Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Memory Based Online Learning of Deep Representations from Video Streams (2017, methodology): “Wefurther compare the proposed algorithm with other state-of-the-art MOT trackers, including modiﬁed versions of TLD [65], ADMM [60], IHTLS [61].”
3D Multi-Object Tracking: A Baseline and New Evaluation Metrics (2019, methodology): “Modern methods often use the motion model [19], [25], [26] and the appearance feature [25], [27], [28].”
The Unmanned Aerial Vehicle Benchmark: Object Detection and Tracking (2018, methodology): “Therefore, we evaluate 8 recent algorithms including online methods (CMOT [2], MDP [50], SORT [6] and DSORT [48]) and batch based methods (GOG [35], CEM [30], SMOT [13] and IOUT [7]).”
Multiple object tracking: A literature review (2014, background): “Notably, specific solutions for sport scenarios [5], [6], aerial scenes [44], [66], generic objects [8], [65], [67], [68], etc.”
Multi-animal pose estimation, identification and tracking with DeepLabCut (2022, background): “On the other hand, tracklets belonging to different tracks would require a higher-order (that is, more complex) model to explain their spatial evolution 48 .”
DETRAC: A New Benchmark and Protocol for Multi-Object Tracking (2015, methodology): “IHTLS [12] Target association formulated as a generalized linear assignment of tracklets incrementally based on the motion dynamics.”
MOT16: A Benchmark for Multi-Object Tracking (2016, background): “The Similar Multi-Object Tracking ( SMOT ) approach [10] speciﬁcally targets situations where target appearance is ambiguous and rather concentrates on using the motion as a primary cue for data association.”
Deep Affinity Network for Multiple Object Tracking (2018, unknown): “7 CompACT [93]+IHTLS [95] 11.”
FairMOT: On the Fairness of Detection and Re-identification in Multiple Object Tracking (2020, background): “Multi-object tracking can be classified into online methods [1], [24], [25], [26], [27] and batch methods [28], [29], [30], [31], [32], [33] based on whether they rely on future frames.”
Simple online and realtime tracking (2016, methodology): “Additionally, methods which inspired this approach ( TBD [20], ALEx-TRAC [5], and SMOT [1]) are also listed.”
Gesture Modeling by Hanklet-Based Hidden Markov Model (2014, methodology): “Hankel matrices have been successfully used in previous works on action recognition [41], tracking [43] and dynamic textures [42].”
MOTChallenge 2015: Towards a Benchmark for Multi-Target Tracking (2015, background): “The Similar Multi-Object Tracking ( SMOT ) approach [15] speciﬁcally targets situations where target appearance is ambiguous and rather concentrates on using the motion as a primary cue for data association.”
Tracking the Untrackable: Learning to Track Multiple Cues with Long-Term Dependencies (2017, background): “n MTT [7, 47, 63, 82, 53]. However,therearemanycaseslinearmotionmodelscannotdeal with long-term occlusions; to remedy this, non-linear motion models are proposed to produce a more accurate prediction [78, 79, 10]. We present a Long Short-Term Memory (LSTM) model which learns to predict similar motio”