Filters








208,797 Hits in 5.5 sec

Multi-Source Domain Adaptation for Object Detection [article]

Xingxu Yao, Sicheng Zhao, Pengfei Xu, Jufeng Yang
2021 arXiv   pre-print
To reduce annotation labor associated with object detection, an increasing number of studies focus on transferring the learned knowledge from a labeled source domain to another unlabeled target domain.  ...  First, we propose a hierarchical feature alignment strategy to conduct strong and weak alignments for low- and high-level features, respectively, considering their different effects for object detection  ...  Due to these domain discrepan-Labeled source domain Sk Unlabeled target domain T Labeled source domain Sj Figure 1 . An example of domain shift in the multi-source scenario for object detection.  ... 
arXiv:2106.15793v1 fatcat:7bq2o3g6t5h73hd4oriwymgfem

Target-Relevant Knowledge Preservation for Multi-Source Domain Adaptive Object Detection [article]

Jiaxi Wu, Jiaxin Chen, Mengzhe He, Yiru Wang, Bo Li, Bingqi Ma, Weihao Gan, Wei Wu, Yali Wang, Di Huang
2022 arXiv   pre-print
Domain adaptive object detection (DAOD) is a promising way to alleviate performance drop of detectors in new scenes.  ...  By this means, the teacher network is enforced to capture target-relevant knowledge, thus benefiting decreasing domain shift when mentoring object detection in the target domain.  ...  Conclusion In this paper, we present a novel multi-source domain adaptation approach for object detection.  ... 
arXiv:2204.07964v1 fatcat:f5wogrhjargn7fbg45wdboco4q

Adaptive Object Detection with Dual Multi-Label Prediction [article]

Zhen Zhao, Yuhong Guo, Haifeng Shen, Jieping Ye
2020 arXiv   pre-print
In this paper, we propose a novel end-to-end unsupervised deep domain adaptation model for adaptive object detection by exploiting multi-label object recognition as a dual auxiliary task.  ...  object category discoveries between the object recognition task and the object detection task.  ...  model is very suitable for adaptive multi-object detection.  ... 
arXiv:2003.12943v2 fatcat:gqfqbagxevgilgwziczfqtfaxa

Multi-level Domain Adaptive learning for Cross-Domain Detection [article]

Rongchang Xie, Fei Yu, Jiachao Wang, Yizhou Wang, Li Zhang
2019 arXiv   pre-print
To solve this problem, we propose a multi-level domain adaptive model to simultaneously align the distributions of local-level features and global-level features.  ...  In recent years, object detection has shown impressive results using supervised deep learning, but it remains challenging in a cross-domain environment.  ...  Domain adaptation for object detection Although domain adaptation has been studied for a long time in classification tasks, its application in object detection is still in its early stages.  ... 
arXiv:1907.11484v2 fatcat:sdj37szt35bpzcj3m4ohsu2cxy

Multi-Level Domain Adaptive Learning for Cross-Domain Detection

Rongchang Xie, Fei Yu, Jiachao Wang, Yizhou Wang, Li Zhang
2019 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)  
To solve this problem, we propose a multi-level domain adaptive model to simultaneously align the distributions of local-level features and global-level features.  ...  In recent years, object detection has shown impressive results using supervised deep learning, but it remains challenging in a cross-domain environment.  ...  Domain adaptation for object detection Although domain adaptation has been studied for a long time in classification tasks, its application in object detection is still in its early stages.  ... 
doi:10.1109/iccvw.2019.00401 dblp:conf/iccvw/XieYWWZ19 fatcat:ahytgtpfyfe4lbql25nzjrgem4

Deep Domain Adaptive Object Detection: a Survey [article]

Wanyi Li, Fuyu Li, Yongkang Luo, Peng Wang, Jia sun
2020 arXiv   pre-print
This paper aims to review the state-of-the-art progress on deep domain adaptive object detection approaches. Firstly, we introduce briefly the basic concepts of deep domain adaptation.  ...  Deep domain adaptive object detection (DDAOD) has emerged as a new learning paradigm to address the above mentioned challenges.  ...  Domain adaptive Faster RCNN [7] is the first work to deal with the domain adaptation problem for object detection.  ... 
arXiv:2002.06797v3 fatcat:mozths3lk5djndue6dzefxuq3q

Unsupervised Subcategory Domain Adaptive Network for 3D Object Detection in LiDAR

Zhiyu Wang, Li Wang, Liang Xiao, Bin Dai
2021 Electronics  
In this paper, we propose a method for object detection using an unsupervised adaptive network, which does not require additional annotation data of the target domain.  ...  We divide the source domain data into different subcategories and use a multi-label discriminator to assign labels dynamically to the target domain data.  ...  for object detection adaptation.  ... 
doi:10.3390/electronics10080927 fatcat:77x323rmmrbpzfg2joophmb27u

Multi-Granularity Alignment Domain Adaptation for Object Detection [article]

Wenzhang Zhou and Dawei Du and Libo Zhang and Tiejian Luo and Yanjun Wu
2022 arXiv   pre-print
Domain adaptive object detection is challenging due to distinctive data distribution between source domain and target domain.  ...  In this paper, we propose a unified multi-granularity alignment based object detection framework towards domain-invariant feature learning.  ...  Unsupervised Domain Adaptation Given the labeled source data and unlabeled target data, unsupervised domain adaptation in object detection attracts the interest of researchers.  ... 
arXiv:2203.16897v1 fatcat:t2u56yojcvawdkixeooztifur4

Universal Domain Adaptive Object Detector [article]

Wenxu Shi, Lei Zhang, Weijie Chen, Shiliang Pu
2022 arXiv   pre-print
Universal domain adaptive object detection (UniDAOD)is more challenging than domain adaptive object detection (DAOD) since the label space of the source domain may not be the same as that of the target  ...  in object detection by introducing a new Multi-Label Scale-Aware Adapter to perform individual alignment between the corresponding scale for two domains.  ...  domain adaptive object detection (DAOD).  ... 
arXiv:2207.01756v1 fatcat:6wuzufnrvnf2ph4cjl6klbuxau

Diversify and Match: A Domain Adaptive Representation Learning Paradigm for Object Detection [article]

Taekyung Kim, Minki Jeong, Seunghyeon Kim, Seokeon Choi, Changick Kim
2019 arXiv   pre-print
We introduce a novel unsupervised domain adaptation approach for object detection.  ...  We construct a structured domain adaptation framework for our learning paradigm and introduce a practical way of DD for implementation.  ...  The architecture of our domain adaptation framework for object detection.  ... 
arXiv:1905.05396v1 fatcat:ejwjkxllozbrnjovydl7kk2sjy

Multi-Target Domain Adaptation via Unsupervised Domain Classification for Weather Invariant Object Detection [article]

Ting Sun and Jinlin Chen and Francis Ng
2021 arXiv   pre-print
Object detection is an essential technique for autonomous driving.  ...  We propose a novel unsupervised domain classification method which can be used to generalize single-target domain adaptation methods to multi-target domains, and design a weather-invariant object detector  ...  Proposed method We consider the problem setting of multi-target domain adaptation for object detection, and denote the source domain as S, and the mixed target domain T mix = {T 1 , T 2 , · · · , T N }  ... 
arXiv:2103.13970v1 fatcat:eonqptukbfcudloijs2gkgkbka

Densely Semantic Enhancement for Domain Adaptive Region-free Detectors [article]

Bo Zhang, Tao Chen, Bin Wang, Xiaofeng Wu, Liming Zhang, Jiayuan Fan
2021 arXiv   pre-print
Unsupervised domain adaptive object detection aims to adapt a well-trained detector from its original source domain with rich labeled data to a new target domain with unlabeled data.  ...  Secondly, considering that region-free detectors recognize objects of different scales using multi-scale feature maps, the DSEM encodes both multi-level semantic representations and multi-instance spatial-contextual  ...  the subsequent two parts for encoding multi-scale features as follows. 3) Multi-level Semantic Representations: In the existing adversarial domain adaptive object detection approaches [48] - [52]  ... 
arXiv:2108.13101v1 fatcat:ne6ba2agbngrzcpqxpczs2j6ai

Exploring Thermal Images for Object Detection in Underexposure Regions for Autonomous Driving [article]

Farzeen Munir, Shoaib Azam, Muhammd Aasim Rafique, Ahmad Muqeem Sheri, Moongu Jeon, Witold Pedrycz
2021 arXiv   pre-print
Although object detection in the visible spectrum domain imaging has matured, thermal object detection lacks effectiveness.  ...  This work proposes a domain adaptation framework which employs a style transfer technique for transfer learning from visible spectrum images to thermal images.  ...  In this work, we have proposed a framework based on domain adaptation for thermal object detection by translating the low-level features adopted from a source domain (RGB) to a target domain (thermal).  ... 
arXiv:2006.00821v2 fatcat:ara72qlzjjf3zctrrvotbbttje

Diversify and Match: A Domain Adaptive Representation Learning Paradigm for Object Detection

Taekyung Kim, Minki Jeong, Seunghyeon Kim, Seokeon Choi, Changick Kim
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
We introduce a novel unsupervised domain adaptation approach for object detection.  ...  We construct a structured domain adaptation framework for our learning paradigm and introduce a practical way of DD for implementation.  ...  The architecture of our domain adaptation framework for object detection.  ... 
doi:10.1109/cvpr.2019.01274 dblp:conf/cvpr/KimJKCK19 fatcat:bcdpw7upv5akxh3tql3ux3mh3q

Domain Adaptive YOLO for One-Stage Cross-Domain Detection [article]

Shizhao Zhang, Hongya Tuo, Jian Hu, Zhongliang Jing
2021 arXiv   pre-print
Domain shift is a major challenge for object detectors to generalize well to real world applications. Emerging techniques of domain adaptation for two-stage detectors help to tackle this problem.  ...  Multi-scale instance level features alignment is presented to reduce instance domain shift effectively , such as variations in object appearance and viewpoint.  ...  Domain Adaptation for Object detection: Adversarial domain adaptation for object detection is explored by Domain Adaptive Faster R-CNN Chen et al. (2018), which uses a two-stage detector Faster R-CNN.  ... 
arXiv:2106.13939v2 fatcat:yipwpyvblrafxeanjbe5papspu
« Previous Showing results 1 — 15 out of 208,797 results