Pseudo-labeling for Scalable 3D Object Detection
| Title: | Pseudo-labeling for Scalable 3D Object Detection |
|---|---|
| Authors: | Caine, Benjamin, Roelofs, Rebecca, Vasudevan, Vijay, Ngiam, Jiquan, Chai, Yuning, Chen, Zhifeng, Shlens, Jonathon |
| Publication Year: | 2021 |
| Collection: | Computer Science |
| Subject Terms: | Computer Science - Computer Vision and Pattern Recognition, Computer Science - Machine Learning |
| More Details: | To safely deploy autonomous vehicles, onboard perception systems must work reliably at high accuracy across a diverse set of environments and geographies. One of the most common techniques to improve the efficacy of such systems in new domains involves collecting large labeled datasets, but such datasets can be extremely costly to obtain, especially if each new deployment geography requires additional data with expensive 3D bounding box annotations. We demonstrate that pseudo-labeling for 3D object detection is an effective way to exploit less expensive and more widely available unlabeled data, and can lead to performance gains across various architectures, data augmentation strategies, and sizes of the labeled dataset. Overall, we show that better teacher models lead to better student models, and that we can distill expensive teachers into efficient, simple students. Specifically, we demonstrate that pseudo-label-trained student models can outperform supervised models trained on 3-10 times the amount of labeled examples. Using PointPillars [24], a two-year-old architecture, as our student model, we are able to achieve state of the art accuracy simply by leveraging large quantities of pseudo-labeled data. Lastly, we show that these student models generalize better than supervised models to a new domain in which we only have unlabeled data, making pseudo-label training an effective form of unsupervised domain adaptation. |
| Document Type: | Working Paper |
| Access URL: | http://arxiv.org/abs/2103.02093 |
| Accession Number: | edsarx.2103.02093 |
| Database: | arXiv |
| FullText | Text: Availability: 0 CustomLinks: – Url: http://arxiv.org/abs/2103.02093 Name: EDS - Arxiv Category: fullText Text: View this record from Arxiv MouseOverText: View this record from Arxiv – Url: https://resolver.ebsco.com/c/xy5jbn/result?sid=EBSCO:edsarx&genre=article&issn=&ISBN=&volume=&issue=&date=20210302&spage=&pages=&title=Pseudo-labeling for Scalable 3D Object Detection&atitle=Pseudo-labeling%20for%20Scalable%203D%20Object%20Detection&aulast=Caine%2C%20Benjamin&id=DOI: Name: Full Text Finder (for New FTF UI) (s8985755) Category: fullText Text: Find It @ SCU Libraries MouseOverText: Find It @ SCU Libraries |
|---|---|
| Header | DbId: edsarx DbLabel: arXiv An: edsarx.2103.02093 RelevancyScore: 1009 AccessLevel: 3 PubType: Report PubTypeId: report PreciseRelevancyScore: 1008.97308349609 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Pseudo-labeling for Scalable 3D Object Detection – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Caine%2C+Benjamin%22">Caine, Benjamin</searchLink><br /><searchLink fieldCode="AR" term="%22Roelofs%2C+Rebecca%22">Roelofs, Rebecca</searchLink><br /><searchLink fieldCode="AR" term="%22Vasudevan%2C+Vijay%22">Vasudevan, Vijay</searchLink><br /><searchLink fieldCode="AR" term="%22Ngiam%2C+Jiquan%22">Ngiam, Jiquan</searchLink><br /><searchLink fieldCode="AR" term="%22Chai%2C+Yuning%22">Chai, Yuning</searchLink><br /><searchLink fieldCode="AR" term="%22Chen%2C+Zhifeng%22">Chen, Zhifeng</searchLink><br /><searchLink fieldCode="AR" term="%22Shlens%2C+Jonathon%22">Shlens, Jonathon</searchLink> – Name: DatePubCY Label: Publication Year Group: Date Data: 2021 – Name: Subset Label: Collection Group: HoldingsInfo Data: Computer Science – Name: Subject Label: Subject Terms Group: Su Data: <searchLink fieldCode="DE" term="%22Computer+Science+-+Computer+Vision+and+Pattern+Recognition%22">Computer Science - Computer Vision and Pattern Recognition</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+Science+-+Machine+Learning%22">Computer Science - Machine Learning</searchLink> – Name: Abstract Label: Description Group: Ab Data: To safely deploy autonomous vehicles, onboard perception systems must work reliably at high accuracy across a diverse set of environments and geographies. One of the most common techniques to improve the efficacy of such systems in new domains involves collecting large labeled datasets, but such datasets can be extremely costly to obtain, especially if each new deployment geography requires additional data with expensive 3D bounding box annotations. We demonstrate that pseudo-labeling for 3D object detection is an effective way to exploit less expensive and more widely available unlabeled data, and can lead to performance gains across various architectures, data augmentation strategies, and sizes of the labeled dataset. Overall, we show that better teacher models lead to better student models, and that we can distill expensive teachers into efficient, simple students. Specifically, we demonstrate that pseudo-label-trained student models can outperform supervised models trained on 3-10 times the amount of labeled examples. Using PointPillars [24], a two-year-old architecture, as our student model, we are able to achieve state of the art accuracy simply by leveraging large quantities of pseudo-labeled data. Lastly, we show that these student models generalize better than supervised models to a new domain in which we only have unlabeled data, making pseudo-label training an effective form of unsupervised domain adaptation. – Name: TypeDocument Label: Document Type Group: TypDoc Data: Working Paper – Name: URL Label: Access URL Group: URL Data: <link linkTarget="URL" linkTerm="http://arxiv.org/abs/2103.02093" linkWindow="_blank">http://arxiv.org/abs/2103.02093</link> – Name: AN Label: Accession Number Group: ID Data: edsarx.2103.02093 |
| PLink | https://login.libproxy.scu.edu/login?url=https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&scope=site&db=edsarx&AN=edsarx.2103.02093 |
| RecordInfo | BibRecord: BibEntity: Subjects: – SubjectFull: Computer Science - Computer Vision and Pattern Recognition Type: general – SubjectFull: Computer Science - Machine Learning Type: general Titles: – TitleFull: Pseudo-labeling for Scalable 3D Object Detection Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Caine, Benjamin – PersonEntity: Name: NameFull: Roelofs, Rebecca – PersonEntity: Name: NameFull: Vasudevan, Vijay – PersonEntity: Name: NameFull: Ngiam, Jiquan – PersonEntity: Name: NameFull: Chai, Yuning – PersonEntity: Name: NameFull: Chen, Zhifeng – PersonEntity: Name: NameFull: Shlens, Jonathon IsPartOfRelationships: – BibEntity: Dates: – D: 02 M: 03 Type: published Y: 2021 |
| ResultId | 1 |