Classifier adaptation at prediction time
The DOI matches its Crossref record; 5 of the 5 references it cites were found in a registry or an index.
Abstract
Classifiers for object categorization are usually evaluated by their accuracy on a set of i.i.d. test examples. This provides us with an estimate of the expected error when applying the classifiers to a single new image. In real application, however, classifiers are rarely only used for a single image and then discarded. Instead, they are applied sequentially to many images, and these are typically not i.i.d. samples from a fixed data distribution, but they carry dependencies and their class distribution varies over time. In this work, we argue that the phenomenon of correlated data at prediction time is not a nuisance, but a blessing in disguise. We describe a probabilistic method for adapting classifiers at prediction time without having to retrain them. We also introduce a framework for creating realistically distributed image sequences, which offers a way to benchmark classifier adaptation methods, such as the one we propose. Experiments on the ILSVRC2010 and ILSVRC2012 datasets show that adapting object classification systems at prediction time can significantly reduce their error rate, even with no additional human feedback.
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
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155
Image Classification with the Fisher Vector: Theory and Practice
Sánchez, J., Perronnin, F., Mensink, T., & Verbeek, J. (2013) · International Journal of Computer Vision · doi:10.1007/s11263-013-0636-x · Cited by 1,498 · cited 1×
“It was trained on 4096 dimensional Fisher vectors [24] using the JSGD toolkit.”
MethodThe devil is in the details: an evaluation of recent feature encoding methods
Chatfield, K., Lempitsky, V., Vedaldi, A., & Zisserman, A. (2011) · Procedings of the British Machine Vision Conference 2011 · doi:10.5244/c.25.76 · Cited by 834 · cited 1×
“In combination with Fisher vector representations they achieve best results amongst those methods that follow the classical computer vision pipeline in which feature extraction and classifier learning are separate steps [4].”
Criminisi, A., Shotton, J., & Konukoglu, E. (2012) · Foundations and Trends® in Computer Graphics and Vision · doi:10.1561/0600000035 · Cited by 677 · cited 1×
“, convolutional networks [1], random forests [6] or one-versus-rest support vector machines [14] with Platt scaling [19].”
MethodAdjusting the Outputs of a Classifier to New a Priori Probabilities: A Simple Procedure
Saerens, M., Latinne, P., & Decaestecker, C. (2002) · Neural Computation · doi:10.1162/089976602753284446 · Cited by 361 · cited 2×
“If π were known, we could immediately derive an optimal adaptation rule in closed form, see [23] for a detailed derivation.”
BackgroundMethod
Cited by
iCaRL: Incremental Classifier and Representation Learning
Rebuffi, S.-A., Kolesnikov, A., Sperl, G., & Lampert, C. H. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.587 · Cited by 3,988
“Royer and Lampert [25] adapt classifiers to a time-varying data stream but their method cannot handle newly appearing classes.”
MethodSelf-citationMemory Aware Synapses: Learning What (not) to Forget
Aljundi, R., Babiloni, F., Elhoseiny, M., Rohrbach, M., & Tuytelaars, T. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01219-9_9 · Cited by 1,590
Rudd, E. M., Jain, L. P., Scheirer, W. J., & Boult, T. E. (2018) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2017.2707495 · Cited by 326
Transfer adaptation learning: A decade survey
Zhang, L., & Gao, X. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3183326 · Cited by 314
Unsupervised Domain Adaptation of Object Detectors: A Survey
Oza, P., Sindagi, V. A., VS, V., & Patel, V. M. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3217046 · Cited by 313
Metadata
- Authors
- Royer, A., & Lampert, C. H.
- Venue
- 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2015-06
- Pages
- 1401-1409
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W1943830695
- Cited by
- 52
- References
- 5
- FWCI
- 3.22
- Type
- conference paper
- Language
- en
- Open access
- closed
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.