Tracking Without Bells and Whistles

By Philipp Bergmann, Tim Meinhardt, L. Leal-Taix'e. IEEE International Conference on Computer Vision, 2019

1,056 citations recorded by the source corpus.

The problem of tracking multiple objects in a video sequence poses several challenging tasks. For tracking-by-detection, these include object re-identification, motion prediction and dealing with occlusions. We present a tracker (without bells and whistles) that accomplishes tracking without specifically targeting any of these tasks, in particular, we perform no training or optimization on tracking data. To this end, we exploit the bounding box regression of an object detector to predict the position of an object in the next frame, thereby converting a detector into a Tracktor. We demonstrate the potential of Tracktor and provide a new state-of-the-art on three multi-object tracking benchmarks by extending it with a straightforward re-identification and camera motion compensation. We then perform an analysis on the performance and failure cases of several state-of-the-art tracking methods in comparison to our Tracktor. Surprisingly, none of the dedicated tracking methods are considerably better in dealing with complex tracking scenarios, namely, small and occluded objects or missing detections. However, our approach tackles most of the easy tracking scenarios. Therefore, we motivate our approach as a new tracking paradigm and point out promising future research directions. Overall, Tracktor yields superior tracking performance than any current tracking method and our analysis exposes remaining and unsolved tracking challenges to inspire future research directions.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/58cf45b4c1bc070535c2f3c1004b2fdc6fe8c907

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

Citations in the corpus

MOTR: End-to-End Multiple-Object Tracking with TRansformer (2021, methodology): “Similarly, Re-ID based approaches [1, 41] compute the feature similarity from adjacent frames and as-”
DanceTrack: Multi-Object Tracking in Uniform Appearance and Diverse Motion (2021, background): “Object tracking has been long studied and can be beneficial to applications such as autonomous driving, video analysis and robot planning [1,4,25,35].”
Observation-Centric SORT: Rethinking SORT for Robust Multi-Object Tracking (2022, methodology): “In addition, we use the commonly adopted detection filtering [4, 80].”
GSM: Graph Similarity Model for Multi-Object Tracking (2020, methodology): “Third, we apply the proposed graph similarity model (GSM) to Tracktor [Bergmann et al. , 2019], the current state-of-the-art online MOT tracker, and achieve the best performance on MOT benchmarks [Milan et al. , 2016] in most metrics including MOTA and IDF1.”
Chained-Tracker: Chaining Paired Attentive Regression Results for End-to-End Joint Multiple-Object Detection and Tracking (2020, methodology): “Online Tracktor+CTdet [7] 54.”
Learning Target Candidate Association to Keep Track of What Not to Track (2021, background): “The dominant paradigm in MOT is tracking-bydetection [5, 1, 49, 40, 52], where tracking is posed as the problem of associating object detections over time.”
StrongSORT: Make DeepSORT Great Again (2022, methodology): “To solve this problem, several recent methods [1,60,74] integrate the detector and embedding model into a unified framework.”
FairMOT: On the Fairness of Detection and Re-identification in Multiple Object Tracking (2020, methodology): “The best-performing MOT methods (Bergmann et al. 2019; Brasó and Leal-Taixé 2020; Hornakova et al. 2020; Yu et al. 2016; Mahmoudi et al. 2019; Zhou et al. 2018; Wojke et al. 2017; Chen et al. 2018a; Wang et al. 2020b; Voigtlaender et al. 2019) usually follow the tracking-by-detection”
Tracking Objects as Points (2020, methodology): “Recent work on simultaneous detection and tracking [1,8] has made progress in alleviating some of this complexity.”
Learning a Neural Solver for Multiple Object Tracking (2019, methodology): “After that, we discard the pedestrian ID assigned by [2], and simply treat the resulting boxes as raw detections for our neural solver.”
Towards Real-Time Multi-Object Tracking (2019, background): “The third category attempts to explore novel MOT paradigms, for instance, incorporating single object trackers into the detector by predicting the spatial offsets (Bergmann, Meinhardt, and Leal-Taixe 2019).”
Opening up Open World Tracking (2021, methodology): “Tracktor [7] almost doesn’t incorrectly merge objects at all (high A.”
TransTrack: Multiple-Object Tracking with Transformer (2020, methodology): “We introduce track rebirth in the inference procedure of TransTrack in order to enhance robustness to occlusions and short-term disappearing [1, 45, 27].”
BoT-SORT: Robust Associations Multi-Pedestrian Tracking (2022, methodology): “Therefore, many researchers adopted camera motion compensation (CMC) [1, 21, 18, 40, 13] by aligning frames via image registration using the Enhanced Correlation Coefﬁcient (ECC) maximization [15] or matching features such as ORB [36].”