A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Filters
Uncertainty-Aware Unsupervised Domain Adaptation in Object Detection
[article]
2021
arXiv
pre-print
Preserved Fulltext
Other Versions
Unsupervised domain adaptive object detection aims to adapt detectors from a labelled source domain to an unlabelled target domain. ...
To address this issue, we design an uncertainty-aware domain adaptation network (UaDAN) that introduces conditional adversarial learning to align well-aligned and poorly-aligned samples separately in different ...
CONCLUSION This paper presents an uncertainty-aware domain adaption technique for unsupervised domain adaptation in object detection. ...
arXiv:2103.00236v2
fatcat:zrr6pjdx55br3lphxfychkifz4
Uncertainty-Aware Model Adaptation for Unsupervised Cross-Domain Object Detection
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
We propose an uncertainty-aware model adaptation method, which is based on two motivations: 1) the estimation and exploitation of model uncertainty in a new domain is critical for reliable domain adaptation ...
This work tackles the unsupervised cross-domain object detection problem which aims to generalize a pre-trained object detector to a new target domain without labels. ...
In this work, we tackle the unsupervised domain adaption problem for object detection. ...
arXiv:2108.12612v1
fatcat:3nbokgoi5bge7k5uvriko73guq
Uncertainty-aware Mean Teacher for Source-free Unsupervised Domain Adaptive 3D Object Detection
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Pseudo-label based self training approaches are a popular method for source-free unsupervised domain adaptation. ...
In order to avoid reinforcing errors caused by label noise, we propose an uncertainty-aware mean teacher framework which implicitly filters incorrect pseudo-labels during training. ...
We propose a framework for unsupervised domain adaptation for 3D object detection. ...
arXiv:2109.14651v1
fatcat:5wo4dotrb5ejtnqcgqneagnkyi
Category Contrast for Unsupervised Domain Adaptation in Visual Tasks
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Other Versions
In this work, we explore the idea of instance contrastive learning in unsupervised domain adaptation (UDA) and propose a novel Category Contrast technique (CaCo) that introduces semantic priors on top ...
Instance contrast for unsupervised representation learning has achieved great success in recent years. ...
Unsupervised domain adaptation aims to leverage unlabelled target data to improve network performance in target domain. ...
arXiv:2106.02885v3
fatcat:6nthlfqbujhgfdlhye4ufmmyne
Table of Contents
2022
2022 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Nasrabadi Unsupervised BatchNorm Adaptation (UBNA): A Domain Adaptation Method for Semantic (West Virginia University) Segmentation Without Using Source Domain Representations OTB-Morph: One-Time Biometrics ...
Aware Proposal Segmentation for Unknown Object Detection Explainable & Interpretable Artificial Intelligence for Biometrics Yimeng Li (George Mason University) and Jana Košecká (George Mason University ...
Class-Aware Object Counting Andreas Michel (Fraunhofer IOSB) ...
doi:10.1109/wacvw54805.2022.00004
fatcat:u376nbmesjc75ontakvwkxwivu
Adverse Weather Image Translation with Asymmetric and Uncertainty-aware GAN
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Finally, we propose uncertainty-aware cycle-consistency loss to address the regional uncertainty of a cyclic reconstructed image. ...
Adverse weather image translation belongs to the unsupervised image-to-image (I2I) translation task which aims to transfer adverse condition domain (eg, rainy night) to standard domain (eg, day). ...
This concept has influenced several unsupervised domain translation tasks such as face attribute editing [5, 11, 19, 24] or domain adaptation [23, 29, 34] . ...
arXiv:2112.04283v3
fatcat:leflugjhufcuxm5uu3xcb6b5ci
Harmonizing Transferability and Discriminability for Adapting Object Detectors
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Recent advances in adaptive object detection have achieved compelling results in virtue of adversarial feature adaptation to mitigate the distributional shifts along the detection pipeline. ...
Moreover, transferability and discriminability may come at a contradiction in adversarial adaptation given the complex combinations of objects and the differentiated scene layouts between domains. ...
UDA for Object Detection By contrast, there is relatively limited study on domain adaptation for object detection task, despite the impressive performance on single domain detection [42, 32, 41, 30, 39 ...
arXiv:2003.06297v1
fatcat:gketkjfr6rgk3b4y4s5nme6jvm
Model Adaptation: Historical Contrastive Learning for Unsupervised Domain Adaptation without Source Data
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Unsupervised domain adaptation aims to align a labeled source domain and an unlabeled target domain, but it requires to access the source data which often raises concerns in data privacy, data portability ...
We study unsupervised model adaptation (UMA), or called Unsupervised Domain Adaptation without Source Data, an alternative setting that aims to adapt source-trained models towards target distributions ...
Uncertainty-aware unsupervised
domain adaptation in object detection. ...
arXiv:2110.03374v5
fatcat:s5ppiflgxbdgrogxxrhtwzuq7q
Uncertainty-Aware Adaptation for Self-Supervised 3D Human Pose Estimation
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
To this end, we cast the 3D human pose learning as an unsupervised domain adaptation problem. ...
Alongside synthetic-to-real 3D pose adaptation, the joint-uncertainties allow expanding the adaptation to work on in-the-wild images even in the presence of occlusion and truncation scenarios. ...
We show uncertainty-aware 3D pose estimation results for unsupervised adaptation to in-the-wild samples with partial body visibility. ...
arXiv:2203.15293v1
fatcat:4n7wi3wolraite52hwnkkmoz34
ST3D++: Denoised Self-training for Unsupervised Domain Adaptation on 3D Object Detection
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
In this paper, we present a self-training method, named ST3D++, with a holistic pseudo label denoising pipeline for unsupervised domain adaptation on 3D object detection. ...
First, ST3D++ pre-trains the 3D object detector on the labeled source domain with random object scaling (ROS) which is designed to reduce target domain pseudo label noise arising from object scale bias ...
This task is also known as unsupervised domain adaptation (UDA) for 3D object detection. ...
arXiv:2108.06682v1
fatcat:nhpe3pvcufeabhtme5ou2fvxxi
A Survey of Visual Sensory Anomaly Detection
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Visual sensory anomaly detection (AD) is an essential problem in computer vision, which is gaining momentum recently thanks to the development of AI for good. ...
Compared with semantic anomaly detection which detects anomaly at the label level (semantic shift), visual sensory AD detects the abnormal part of the sample (covariate shift). ...
To handle this problem, A-SVDD [Sindagi and Srivastava, 2017] uses domain adaptation in unsupervised and semi-supervised settings to learn an incremental classifier based on the existing support vector ...
arXiv:2202.07006v1
fatcat:2bqzmmrnjzggti5tcewa3mh3sa
ST3D: Self-training for Unsupervised Domain Adaptation on 3D Object Detection
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
We present a new domain adaptive self-training pipeline, named ST3D, for unsupervised domain adaptation on 3D object detection from point clouds. ...
These specific designs for 3D object detection enable the detector to be trained with consistent and high-quality pseudo labels and to avoid overfitting to the large number of easy examples in pseudo labeled ...
This task is also known as unsupervised domain adaptation (UDA) for 3D object detection. ...
arXiv:2103.05346v2
fatcat:kdmjr5yy5vcx3baf3joqj3rj4u
SADANet: Integrating Scale-Aware and Domain Adaptive for Traffic Sign Detection
2020
IEEE Access
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
In this paper, a traffic sign detection framework using scale-aware and domain adaptive network (SADANet) was proposed, which seamlessly combines a multiscale prediction network (MSPN) with a domain adaptive ...
INDEX TERMS Adaptive feature weighting, domain adaptation, multiscale feature, traffic sign detection. 77920 This work is licensed under a Creative Commons Attribution 4.0 License. ...
Domain adaption network (DAN) is dedicated to address the cross-domain object detection problem by unsupervised domain adaptation. ...
doi:10.1109/access.2020.2989758
fatcat:d7smukppfrcldn3thpklwa6bqe
Exploiting Negative Learning for Implicit Pseudo Label Rectification in Source-Free Domain Adaptive Semantic Segmentation
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
unsupervised learning: Maximum squares loss applies to regularize the target model to ensure the confidence in prediction; and 2) Noise-aware pseudo label learning: Negative learning enables tolerance ...
Aiming at these pitfalls, this study develops a domain adaptive solution to semantic segmentation with pseudo label rectification (namely PR-SFDA), which operates in two phases: 1) Confidence-regularized ...
Equation 1 formulates the basic optimization objective for source-present domain adaptation (SDA), where L SRC is supervised loss in source domain, L DA is the adaptation loss in both domain (which comes ...
arXiv:2106.12123v1
fatcat:tgxeuol4vfbbvdbaorrej4nqsm
DyStaB: Unsupervised Object Segmentation via Dynamic-Static Bootstrapping
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
We describe an unsupervised method to detect and segment portions of images of live scenes that, at some point in time, are seen moving as a coherent whole, which we refer to as objects. ...
Our models are compared to the state of the art in both video object segmentation and salient object detection. ...
DAVIS FBMS
Effectiveness of Confidence-Aware Adaptation Here we check the effectiveness of the confidence-aware adaptation scheme proposed in Sec. 3.2 in precluding counter-productive self-learning. ...
arXiv:2008.07012v2
fatcat:ofh5xvyipnak7kaxagz4jlv7ny
« Previous
Showing results 1 — 15 out of 6,010 results