26,738 Hits in 8.7 sec

A Modified Self-training Method for Adapting Domains in the Task of Food Classification

Elnaz Heravi, Hamed Aghdam, Domenec Puig
2019 Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications  
This potentially reduces the performance of the network. In this paper, we propose a method based on self-training to perform unsupervised domain adaptation in the task of food classification.  ...  Food trackers are tools that recognize foods using their images. In the core of these tools there is usually a neural network that performs the classification.  ...  Modified Self-training Next, we utilized our modified self-training approach for adapting the domain from the Food-101 to the UPMC-101.  ... 
doi:10.5220/0007688801430154 dblp:conf/visapp/HeraviAP19 fatcat:36fcv2wblfghrh6kqwp5tgq6ou

Use of Transfer Learning for Automatic Dietary Monitoring through Throat Microphone Recordings

Mehmet Ali Tugtekin Turan, Engin Erzin
2019 Zenodo  
We propose a new domain adaptation framework in a heterogeneous setup based on T/S learning paradigm.  ...  This allows the use of a significantly larger set of adaptation data, adds robustness to the resulting model, and significantly improves the performance of food in [...]  ...  One of the goals in this thesis is to develop non-invasive monitoring and automated methods for the classification of food intake.  ... 
doi:10.5281/zenodo.3841957 fatcat:so4kiaj4ljbw5aay36xd6dlx2q

Use of Transfer Learning for Automatic Dietary Monitoring through Throat Microphone Recordings

Mehmet Ali Tugtekin Turan, Engin Erzin
2019 Zenodo  
We propose a new domain adaptation framework in a heterogeneous setup based on T/S learning paradigm.  ...  This allows the use of a significantly larger set of adaptation data, adds robustness to the resulting model, and significantly improves the performance of food in [...]  ...  One of the goals in this thesis is to develop non-invasive monitoring and automated methods for the classification of food intake.  ... 
doi:10.5281/zenodo.3841956 fatcat:ncalroecszg3hhpc45havcxhee

AdaptFormer: Adapting Vision Transformers for Scalable Visual Recognition [article]

Shoufa Chen, Chongjian Ge, Zhan Tong, Jiangliu Wang, Yibing Song, Jue Wang, Ping Luo
2022 arXiv   pre-print
Although the pre-trained Vision Transformers (ViTs) achieved great success in computer vision, adapting a ViT to various image and video tasks is challenging because of its heavy computation and storage  ...  To address this challenge, we propose an effective adaptation approach for Transformer, namely AdaptFormer, which can adapt the pre-trained ViTs into many different image and video tasks efficiently.  ...  In summary, we present a strategy for tuning a pre-trained vision Transformer on a set of scalable vision recognition tasks (e.g.image domain and video domain).  ... 
arXiv:2205.13535v1 fatcat:7mlu63m7dnctpjao7cyhhju7mi

Practical Evaluation of Out-of-Distribution Detection Methods for Image Classification [article]

Engkarat Techapanurak, Takayuki Okatani
2021 arXiv   pre-print
While the unified evaluation method is necessary for a fair comparison, there is a question of if its choice of tasks and datasets reflect real-world applications and if the evaluation results can generalize  ...  Our results can also be used as a guide for practitioners for the selection of OOD detection methods.  ...  It is to make a model that has learned a task using the dataset of a particular domain adapt to work on data from a different domain.  ... 
arXiv:2101.02447v1 fatcat:7rmalvpjdjbhblfdjyfhor2bee

Neuromodulated Dopamine Plastic Networks for Heterogeneous Transfer Learning with Hebbian Principle

Arjun Magotra, Juntae Kim
2021 Symmetry  
These activities are controlled by neuromodulation, which is itself under the control of the brain. The subjective brain's self-modifying abilities play an essential role in learning and adaptation.  ...  The artificial neural networks with neuromodulated plasticity are used to implement transfer learning in the image classification domain.  ...  The brain's resulting self-modifying abilities play an important role in learning and adaptation and constitute a significant basis for biological memorization and learning sustainabil-ity.  ... 
doi:10.3390/sym13081344 fatcat:mj4nxsut6vgojd266qfuuhfrlu

MixNorm: Test-Time Adaptation Through Online Normalization Estimation [article]

Xuefeng Hu, Gokhan Uzunbas, Sirius Chen, Rui Wang, Ashish Shah, Ram Nevatia, Ser-Nam Lim
2021 arXiv   pre-print
The proposed method significantly outperforms the State-Of-The-Art in the newly proposed settings in Test-Time Adaptation Task, and also demonstrates improvements in various other settings such as Source-Free  ...  Unsupervised Domain Adaptation and Zero-Shot Classification.  ...  In addition to the Test-Time Adaptation task, the proposed MixNorm Layer demonstrates improvement over two other tasks, source-free Domain Adaptation and Zero-Shot Image Classification.  ... 
arXiv:2110.11478v1 fatcat:po2tg35wpfciroi434rv6gkrx4

DILBERT: Customized Pre-Training for Domain Adaptation withCategory Shift, with an Application to Aspect Extraction [article]

Entony Lekhtman, Yftah Ziser, Roi Reichart
2021 arXiv   pre-print
Particularly, fine-tuning a pre-trained language model on a source domain and then applying it to a different target domain, results in a sharp performance decline of the eventual classifier for many source-target  ...  The rise of pre-trained language models has yielded substantial progress in the vast majority of Natural Language Processing (NLP) tasks.  ...  Acknowledgments We would like to thank the members of the IE@Technion NLP group for their valuable feedback and advice. This research was partially funded by an ISF personal grant No. 1625/18.  ... 
arXiv:2109.00571v1 fatcat:pbz444kh4fcolagl5fjupkluba

Self-Adaptive Forecasting for Improved Deep Learning on Non-Stationary Time-Series [article]

Sercan O. Arik, Nathanael C. Yoder, Tomas Pfister
2022 arXiv   pre-print
In this paper, we propose a novel method, Self-Adaptive Forecasting (SAF), to modify the training of time-series forecasting models to improve their performance on forecasting tasks with such non-stationary  ...  This is a form of test-time training that creates a self-supervised learning problem on test samples before performing the prediction task.  ...  In contrast to learning methods that modify the loss function only during training, such as multi-task learning or autoencoding, test-time training has a loss function for test time (see Fig. 1 ), which  ... 
arXiv:2202.02403v1 fatcat:eqs3y45nczawbooo6y7bxjqv7y

Activation Regression for Continuous Domain Generalization with Applications to Crop Classification [article]

Samar Khanna, Bram Wallace, Kavita Bala, Bharath Hariharan
2022 arXiv   pre-print
We develop a dataset spatially distributed across the entire continental United States, providing macroscopic insight into the effects of geography on crop classification in multi-spectral and temporally  ...  In this paper, we model geographic generalisation in medium resolution Landsat-8 satellite imagery as a continuous domain adaptation problem, demonstrating how models generalise better with appropriate  ...  Moreover, recent work in self-supervised methods for satellite imagery [7] is relevant to the question of generalising to out-of-domain tasks and distributions, and may work in tandem with domain generalisation  ... 
arXiv:2204.07030v1 fatcat:kbu5fnyd45gmjnhdux5jl45rp4

Semi-Supervised Learning for Fine-Grained Classification with Self-Training

Obed Tettey Nartey, Guowu Yang, Jinzhao Wu, Sarpong Kwadwo Asare
2019 IEEE Access  
In this paper we adopt a semi-supervised self-training method to increase the amount of training data, prevent overfitting and improve the performance of deep models by proposing a novel selection algorithm  ...  The method improves the accuracy in two-fold; bridging the gap in the appearance of visual objects, and enlarging the training set to meet the demands of deep models.  ...  The authors would like to thank the anonymous reviewers for their careful reading of this article and for their helpful and constructive comments.  ... 
doi:10.1109/access.2019.2962258 fatcat:gmduhkjdvne2tmvtd6v276nfqm

Benchmarking Representation Learning for Natural World Image Collections

Grant Van Horn, Elijah Cole, Sara Beery, Kimberly Wilber, Serge Belongie, Oisin MacAodha
2021 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
We designed the latter, NeWT, in collaboration with domain experts with the aim of benchmarking the performance of representation learning algorithms on a suite of challenging natural world binary classification  ...  a diverse set of tasks.  ...  Acknowledgments Thanks to the iNaturalist team and community for providing access to data, Eliot Miller and Mitch Barry for helping to curate NeWT, and to Pietro Perona for valuable feedback.  ... 
doi:10.1109/cvpr46437.2021.01269 fatcat:gopl7tljkrgavekagz2b7fjl5e

Generative Adversarial Networks for Image Augmentation in Agriculture: A Systematic Review [article]

Ebenezer Olaniyi, Dong Chen, Yuzhen Lu, Yanbo Huang
2022 arXiv   pre-print
In agricultural image analysis, optimal model performance is keenly pursued for better fulfilling visual recognition tasks (e.g., image classification, segmentation, object detection and localization),  ...  Since 2017, there has been a growth of research into GANs for image augmentation or synthesis in agriculture for improved model performance.  ...  Acknowledgement This work was supported in part by Cotton Incorporated award #21-005 and the USDA National Institute of Food and Agriculture Hatch project #1025922.  ... 
arXiv:2204.04707v2 fatcat:wcvmq3vl35fo7on2pqyblbzcku

Convolutional Bypasses Are Better Vision Transformer Adapters [article]

Shibo Jie, Zhi-Hong Deng
2022 arXiv   pre-print
Different from other PETL methods, Convpass benefits from the hard-coded inductive bias of convolutional layers and thus is more suitable for visual tasks, especially in the low-data regime.  ...  In this paper, we propose to construct Convolutional Bypasses (Convpass) in ViT as adaptation modules, introducing only a small amount (less than 0.5% of model parameters) of trainable parameters to adapt  ...  For our methods, we train the Convpass modules and classification heads for 50 epochs. Other hyperparameters are listed in the Appendix. Results The results are shown in Table 2 .  ... 
arXiv:2207.07039v3 fatcat:7l2iqgb35bbv5nb7v34fbkrira

Can domain adaptation make object recognition work for everyone? [article]

Viraj Prabhu, Ramprasaath R. Selvaraju, Judy Hoffman, Nikhil Naik
2022 arXiv   pre-print
We demonstrate the inefficacy of standard DA methods at Geographical DA, highlighting the need for specialized geographical adaptation solutions to address the challenge of making object recognition work  ...  We investigate the effectiveness of unsupervised domain adaptation (UDA) of such models across geographies at closing this performance gap.  ...  Per-category target test accuracy for a model trained on blue: target train set and orange: source train set and adapted to the target domain via the SST method.  ... 
arXiv:2204.11122v1 fatcat:auaqbcspgbegfp6v7j2svnacti
« Previous Showing results 1 — 15 out of 26,738 results