Filters








1,901 Hits in 5.7 sec

From ImageNet to Image Classification: Contextualizing Progress on Benchmarks [article]

Dimitris Tsipras, Shibani Santurkar, Logan Engstrom, Andrew Ilyas, Aleksander Madry
2020 arXiv   pre-print
In this work, we use human studies to investigate the consequences of employing such a pipeline, focusing on the popular ImageNet dataset.  ...  Our analysis pinpoints how a noisy data collection pipeline can lead to a systematic misalignment between the resulting benchmark and the real-world task it serves as a proxy for.  ...  ., via simple crowd-sourcing) whether further progress on the ImageNet benchmark is meaningful, or is simply a result of overfitting to this benchmarks' idiosyncrasies.  ... 
arXiv:2005.11295v1 fatcat:ablcgpxjvrfs5gulctw4nessem

Towards Non-I.I.D. Image Classification: A Dataset and Baselines [article]

Yue He, Zheyan Shen, Peng Cui
2019 arXiv   pre-print
Meanwhile, we propose a baseline model with ConvNet structure for General Non-I.I.D. image classification, where distribution of testing data is unknown but different from training data.  ...  In literature, however, the Non-I.I.D. image classification problem is largely understudied. A key reason is lacking of a well-designed dataset to support related research.  ...  Take image classification, a prominent learning task, as an example. Its development benefits a lot from the benchmark datasets, such as PASCAL VOC [7] , MSCOCO [8] , and ImageNet [9] .  ... 
arXiv:1906.02899v3 fatcat:35egblk5sjfkzpwbqfnbmy4hn4

Deep Learning for Scene Classification: A Survey [article]

Delu Zeng, Minyu Liao, Mohammad Tavakolian, Yulan Guo, Bolei Zhou, Dewen Hu, Matti Pietikäinen, Li Liu
2021 arXiv   pre-print
Scene classification, aiming at classifying a scene image to one of the predefined scene categories by comprehending the entire image, is a longstanding, fundamental and challenging problem in computer  ...  big raw data, have been bringing remarkable progress in the field of scene representation and classification.  ...  ACKNOWLEDGMENTS The authors would like to thank the pioneer researchers in scene classification and other related fields. This work was supported in part by grants from National Science  ... 
arXiv:2101.10531v2 fatcat:hwqw5so46ngxdlnfw7zynmpu6m

Satellite Image Scene Classification via ConvNet with Context Aggregation [article]

Zhao Zhou, Yingbin Zheng, Hao Ye, Jian Pu, Gufei Sun
2018 arXiv   pre-print
The ResNet-TP based representation is generated by global average pooling on the last convolutional layers from both pathways.  ...  Scene classification is a fundamental problem to understand the high-resolution remote sensing imagery.  ...  This work was supported in part by grants from National Natural Science Foundation of China (No. 61602459) and Science and Technology Commission of Shanghai Municipality (No. 17511101902 and No. 18511103103  ... 
arXiv:1802.00631v2 fatcat:wdwrdyuisjhtxhwcccu5j4jopu

Contemplating real-world object classification [article]

Ali Borji
2021 arXiv   pre-print
Prior research on this problem has primarily focused on ImageNet variations (e.g., ImageNetV2, ImageNet-A). To avoid potential inherited biases in these studies, we take a different approach.  ...  Deep object recognition models have been very successful over benchmark datasets such as ImageNet.  ...  A large number of studies have tested new ideas by training deep models on ImageNet (from scratch), or by finetuning pre-trained (on ImageNet) classification models on other datasets.  ... 
arXiv:2103.05137v2 fatcat:bu2spm6so5dfnlflrozeq6it3a

Aerial Scene Parsing: From Tile-level Scene Classification to Pixel-wise Semantic Labeling [article]

Yang Long and Gui-Song Xia and Liangpei Zhang and Gong Cheng and Deren Li
2022 arXiv   pre-print
When transferring knowledge from Million-AID, fine-tuning CNN models pretrained on Million-AID perform consistently better than those pretrained ImageNet for aerial scene classification.  ...  In this paper, we address these issues in ASP, with perspectives from tile-level scene classification to pixel-wise semantic labeling.  ...  It is shown that the interpretation prototype of aerial images has been progressing with the improvement of image resolution and experienced the stages from pixel-wise image classification, segmentation-based  ... 
arXiv:2201.01953v2 fatcat:ikigc6f44rfw3eiirijjgdd2pu

Satellite Image Scene Classification via ConvNet With Context Aggregation [chapter]

Zhao Zhou, Yingbin Zheng, Hao Ye, Jian Pu, Gufei Sun
2018 Lecture Notes in Computer Science  
The ResNet-TP based representation is generated by global average pooling on the last convolutional layers from both pathways.  ...  Scene classification is a fundamental problem to understand the high-resolution remote sensing imagery.  ...  This work was supported in part by grants from National Natural Science Foundation of China (No. 61602459) and Science and Technology Commission of Shanghai Municipality (No. 17511101902 and No. 18511103103  ... 
doi:10.1007/978-3-030-00767-6_31 fatcat:egw2q5ignrcdtd5x7lzeg7rdje

SoT: Delving Deeper into Classification Head for Transformer [article]

Jiangtao Xie, Ruiren Zeng, Qilong Wang, Ziqi Zhou, Peihua Li
2021 arXiv   pre-print
For CV tasks, our SoT significantly improves state-of-the-art vision transformers on challenging benchmarks including ImageNet and ImageNet-A.  ...  Despite great advance, most of works only focus on improvement of architectures but pay little attention to the classification head.  ...  Tokens- to-token ViT: Training vision transformers from scratch on imagenet.  ... 
arXiv:2104.10935v2 fatcat:k3bhpgeemjdb7jqliowkhicmya

Remote Sensing Image Scene Classification with Self-Supervised Paradigm under Limited Labeled Samples [article]

Chao Tao, Ji Qi, Weipeng Lu, Hao Wang, Haifeng Li
2020 arXiv   pre-print
Experiments on three commonly used RSIs scene classification datasets demonstrated that this new learning paradigm outperforms the traditional dominant ImageNet pre-trained model.  ...  When labeled samples are not sufficient, the most common solution is to fine-tune the pre-training models using a large natural image dataset (e.g. ImageNet).  ...  Although many researches [5] - [7] have demonstrated that the features extracted from ImageNet pre-trained DCNNs can generalize well to aerial image scene classification tasks, pre-training CNNs on  ... 
arXiv:2010.00882v1 fatcat:xqxmrld7p5avpng67hvbsrraly

Knowledge Graph-Based Image Classification Refinement

Dehai Zhang, Menglong Cui, Yun Yang, Po Yang, Cheng Xie, Di Liu, Beibie Yu, Zhibo Chen
2019 IEEE Access  
We conduct extensive experiments on large-scale image datasets (ImageNet), demonstrating the effectiveness of our approach.  ...  How to make full use of the semantic relationships in categories and how to apply the knowledge of biological vision to image classification are our main concerns.  ...  Compared to GoogLeNet, VGG shows small gains in classification performance on the ImageNet dataset.  ... 
doi:10.1109/access.2019.2912627 fatcat:63ien7u5vfgzrjy6dhyebp56tu

An End-to-End Breast Tumour Classification Model Using Context-Based Patch Modelling- A BiLSTM Approach for Image Classification

Suvidha Tripathi, Satish Kumar Singh, Hwee Kuan Lee
2020 Computerized Medical Imaging and Graphics  
For the given task of classification, we have used BiLSTMs to model both forward and backward contextual relationship.  ...  Researchers working on computational analysis of Whole Slide Images (WSIs) in histopathology have primarily resorted to patch-based modelling due to large resolution of each WSI.  ...  We are also grateful to the NVIDIA corporation for supporting our research in this area by granting us TitanX (PASCAL) GPU.  ... 
doi:10.1016/j.compmedimag.2020.101838 pmid:33340945 fatcat:wgsxhtpijbevzmov6a32kkmgva

HCP: A Flexible CNN Framework for Multi-Label Image Classification

Yunchao Wei, Wei Xia, Min Lin, Junshi Huang, Bingbing Ni, Jian Dong, Yao Zhao, Shuicheng Yan
2016 IEEE Transactions on Pattern Analysis and Machine Intelligence  
Convolutional Neural Network (CNN) has demonstrated promising performance in single-label image classification tasks.  ...  However, how CNN best copes with multi-label images still remains an open problem, mainly due to the complex underlying object layouts and insufficient multi-label training images.  ...  Step2: Fine-tuning on multi-label image set To adapt the pre-trained model on ImageNet to HCP, the entire images from a multi-label image set, e.g., Pascal VOC, are then utilized to further adjust the  ... 
doi:10.1109/tpami.2015.2491929 pmid:26513778 fatcat:jtnufkjow5envpw2d24atnomli

Benchmarking Deep Learning Models for Classification of Book Covers

Adriano Lucieri, Huzaifa Sabir, Shoaib Ahmed Siddiqui, Syed Tahseen Raza Rizvi, Brian Kenji Iwana, Seiichi Uchida, Andreas Dengel, Sheraz Ahmed
2020 SN Computer Science  
Second, it benchmarks the performance on a battery of stateof-the-art image classification models for the task of book cover classification.  ...  While various modalities are available (e.g., cover, title, author, abstract), benchmarking the image-based classification systems based on minimal information is a particularly exciting field due to the  ...  Compliance with ethical standards Conflict of Interest On behalf of all authors, the corresponding author states that there is no conflict of interest.  ... 
doi:10.1007/s42979-020-00132-z fatcat:my6vmfqc7vauzouh7ndqbphlci

DeepPap: Deep Convolutional Networks for Cervical Cell Classification

Ling Zhang, Le Lu, Isabella Nogues, Ronald M. Summers, Shaoxiong Liu, Jianhua Yao
2017 IEEE journal of biomedical and health informatics  
In the testing phase, aggregation is used to average the prediction scores of a similar set of image patches. The proposed method is evaluated on both Pap smear and LBC datasets.  ...  First, the ConvNet is pre-trained on a natural image dataset.  ...  The base ConvNet (denote as ConvNet-B) is pre-trained on the ImageNet classification dataset.  ... 
doi:10.1109/jbhi.2017.2705583 pmid:28541229 fatcat:eox3nd3az5b53nhmrcmgiik3cy

Modeling Multimodal Clues in a Hybrid Deep Learning Framework for Video Classification [article]

Yu-Gang Jiang, Zuxuan Wu, Jinhui Tang, Zechao Li, Xiangyang Xue, Shih-Fu Chang
2017 arXiv   pre-print
Finally, we also propose to refine the prediction scores by leveraging contextual relationships among video semantics.  ...  Experimental results on two challenging benchmarks, the UCF-101 and the Columbia Consumer Videos (CCV), provide strong quantitative evidence that our framework achieves promising results: 93.1% on the  ...  ACKNOWLEDGMENT This work was supported in part by two grants from NSF China (#61622204, #61572134) and two grants from STCSM, Shanghai, China (#16QA1400500, #16JC1420401).  ... 
arXiv:1706.04508v1 fatcat:kn4xvnayifganmwraj5rmlasie
« Previous Showing results 1 — 15 out of 1,901 results