Eliminating Exposure Bias and Metric Mismatch in Multiple Object Tracking
Open accessThe DOI matches its Crossref record; 18 of the 18 references it cites were found in a registry or an index.
Abstract
Identity Switching remains one of the main difficulties Multiple Object Tracking (MOT) algorithms have to deal with. Many state-of-the-art approaches now use sequence models to solve this problem but their training can be affected by biases that decrease their efficiency. In this paper, we introduce a new training procedure that confronts the algorithm to its own mistakes while explicitly attempting to minimize the number of switches, which results in better training. We propose an iterative scheme of building a rich training set and using it to learn a scoring function that is an explicit proxy for the target tracking metric. Whether using only simple geometric features or more sophisticated ones that also take appearance into account, our approach outperforms the state-of-the-art on several MOT benchmarks.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Show and tell: A neural image caption generator
Vinyals, O., Toshev, A., Bengio, S., & Erhan, D. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298935 · Cited by 6,318
Simple online and realtime tracking
Bewley, A., Ge, Z., Ott, L., Ramos, F., & Upcroft, B. (2016) · 2016 IEEE International Conference on Image Processing (ICIP) · doi:10.1109/icip.2016.7533003 · Cited by 4,433
Categorical Reparameterization with Gumbel-Softmax
Jang, E., Gu, S., & Poole, B. (2016) · arXiv · doi:10.48550/arxiv.1611.01144 · Cited by 3,616
Evaluating Multiple Object Tracking Performance: The CLEAR MOT Metrics
Bernardin, K., & Stiefelhagen, R. (2008) · EURASIP Journal on Image and Video Processing · doi:10.1155/2008/246309 · Cited by 2,918 · cited 2×
“This introduces a mismatch with the metric, usually IDF [45] or MOTA [7], which reflects more reliably the desirable behavior of the algorithm.”
BackgroundMethodA Reduction of Imitation Learning and Structured Prediction to No-Regret Online Learning
Ross, S., Gordon, G. J., & Bagnell, J. A. (2018) · Carnegie Mellon University · doi:10.1184/r1/6550949 · Cited by 1,941
Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation
Bengio, Y., Léonard, N., & Courville, A. (2013) · arXiv · doi:10.48550/arxiv.1308.3432 · Cited by 1,910 · cited 2×
“More generally, techniques that allow training models without differentiable relation between inputs and outputs such as policy gradient [52], straight-through estimation [5], and GumbelSoftmax [23] can be seen as methods reducing exposure bias.”
MethodYou'll never walk alone: Modeling social behavior for multi-target tracking
Pellegrini, S., Ess, A., Schindler, K., & van Gool, L. (2009) · 2009 IEEE 12th International Conference on Computer Vision · doi:10.1109/iccv.2009.5459260 · Cited by 1,683 · cited 1×
“Many approaches rely on appearance [6, 17, 28, 57, 58, 12, 32, 46], motion [15], or social cues [20, 43].”
BackgroundScheduled Sampling for Sequence Prediction with Recurrent Neural Networks
Bengio, S., Vinyals, O., Jaitly, N., & Shazeer, N. (2015) · arXiv · doi:10.48550/arxiv.1506.03099 · Cited by 1,621
Sequence Level Training with Recurrent Neural Networks
Ranzato, M., Chopra, S., Auli, M., & Zaremba, W. (2015) · arXiv · doi:10.48550/arxiv.1511.06732 · Cited by 1,426 · cited 1×
“Since exposure bias and metric mismatch (also called loss-evaluation mismatch [49]) are also problems in Natural Language Processing (NLP) [51] and in particular in machine translation [53], several methods have been proposed in these fields to reduce it [44, 4].”
MethodMOT16: A Benchmark for Multi-Object Tracking
Milan, A., Leal-Taixe, L., Reid, I., Roth, S., & Schindler, K. (2016) · arXiv · doi:10.48550/arxiv.1603.00831 · Cited by 1,309 · cited 1×
“Increasing availability of annotated training data and benchmarks, such as MOT15-17 [29, 38], DukeMTMC [45], PathTrack [37], and Wildtrack [10] now makes it possible to learn the data association models required to leverage this knowledge.”
Background
Cited by
HOTA: A Higher Order Metric for Evaluating Multi-object Tracking
Luiten, J., Os̆ep, A., Dendorfer, P., Torr, P., Geiger, A., Leal-Taixé, L., & Leibe, B. (2021) · International Journal of Computer Vision · doi:10.1007/s11263-020-01375-2 · Cited by 1,171
MOTChallenge: A Benchmark for Single-Camera Multiple Target Tracking
Dendorfer, P., Os̆ep, A., Milan, A., Schindler, K., Cremers, D., Reid, I., Roth, S., & Leal-Taixé, L. (2021) · International Journal of Computer Vision · doi:10.1007/s11263-020-01393-0 · Cited by 374
“4 HISP T17 [Baisa, 2019b] 44.62 38.79 77.19 1.43 355 913 25478 276395 10617 7487 208.12 146.76 GMPHD DAL [Baisa, 2019] 44.40 36.23 77.42 1.08 350 927 19170 283380 11137 13900 223.74 279.25 SAS MOT17 [Maksai and Fua, 2019] 44.24 57.18 76.42 1.66 379 1044 29473 283611 1529 2644 30.74 53.16 GMPHD SHA [Song and Jeon, 2016…”
InfluentialGSM: Graph Similarity Model for Multi-Object Tracking
Liu, Q., Chu, Q., Liu, B., & Yu, N. (2020) · Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2020/74 · Cited by 123
“Generally, the existing works can be categorized into online meth-ods [Sadeghian et al. , 2017; Zhu et al. , 2018; Xu et al. , 2019; Bergmann et al. , 2019] and offline methods [Li et al. , 2009; Maksai and Fua, 2019].”
BackgroundProblems and opportunities in training deep learning software systems: an analysis of variance
Pham, H. V., Qian, S., Wang, J., Lutellier, T., Rosenthal, J., Tan, L., Yu, Y., & Nagappan, N. (2020) · Proceedings of the 35th IEEE/ACM International Conference on Automated Software Engineering · doi:10.1145/3324884.3416545 · Cited by 100
“Anecdotal evidence of NI-factors: Some studies [59, 76, 77, 108] quantify the variance in their results caused by NI-factors.”
Background
Metadata
- Authors
- Maksai, A., & Fua, P.
- Venue
- 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2019-06
- Pages
- 4634-4643
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W2948777210
- Cited by
- 67
- References
- 18
- FWCI
- 4.98
- Type
- conference paper
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.