Papers with Code 아카이브

Image Classification

태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.

하위 태스크
33
평가표
177
보존된 평가 기록
2,906

Image Classification is a fundamental task in vision recognition that aims to understand and categorize an image as a whole under a specific label. Unlike object detection, which involves classification and location of multiple objects within an image, image classification typically pertains to single-object images. When the classification becomes highly detailed or reaches instance-level, it is often referred to as image retrieval, which also involves finding similar images in a large database.

Source: Metamorphic Testing for Object Detection Systems

데이터셋

출처에 같은 태스크 이름으로 연결된 데이터셋입니다. 동명 태스크가 포함될 수 있으므로, 비교할 때는 평가 결과의 조건도 확인하세요.

ImageNet-1k vs NINCO

The NINCO (No ImageNet Class Objects) dataset is introduced in the ICML 2023 paper In or Out? Fixing ImageNet Out-of-Distribution Detection Evaluation. The images in this dataset are free from objects that belong to any of the 1000 classes of ImageNet-1K (ILSVRC2012), which makes NINCO suitable for evaluating out-of-distribution detection on ImageNet-1K . The NINCO main dataset consists of 64 OOD classes with a total of 5879 samples. These OOD classes were selected to have no categorical overlap with any classes of ImageNet-1K. Each sample was inspected individually by the authors to not contain ID objects. Besides NINCO, included are (in the same .tar.gz file) truly OOD versions of 11 popular OOD datasets with in total 2715 OOD samples. Further included are 17 OOD unit-tests, with 400 samples each. Code for loading and evaluating on each of the three datasets is provided at https://github.com/j-cb/NINCO. When using NINCO, please consider citing (besides the bibtex given below) the following data sources that were used to create NINCO: Hendrycks et al.: ”Scaling out-of-distribution detection for real-world settings”, ICML, 2022. Bossard et al.: ”Food-101 – mining discriminative components with random forests”, ECCV 2014. Zhou et al.: ”Places: A 10 million image database for scene recognition”, IEEE PAMI 2017. Huang et al.: ”Mos: Towards scaling out-of-distribution detection for large semantic space”, CVPR 2021. Li et al.: ”Caltech 101 (1.0)”, 2022. Ismail et al.: ”MYNursingHome: A fully-labelled image dataset for indoor object classification.”, Data in Brief (V. 32) 2020. The iNaturalist project: https://www.inaturalist.org/ When using NINCO_popular_datasets_subsamples, additionally to the above, please consider citing: Cimpoi et al.: ”Describing textures in the wild”, CVPR 2014. Hendrycks et al.: ”Natural adversarial examples”, CVPR 2021. Wang et al.: ”Vim: Out-of-distribution with virtual-logit matching”, CVPR 2022. Bendale et al.: ”Towards Open Set Deep Networks”, CVPR 2016. Vaze et al.: ”Open-set Recognition: a Good Closed-set Classifier is All You Need?”, ICLR 2022. Wang et al.: ”Partial and Asymmetric Contrastive Learning for Out-of-Distribution Detection in Long-Tailed Recognition.” ICML, 2022. Galil et al.: “A framework for benchmarking Class-out-of-distribution detection and its application to ImageNet”, ICLR 2023.

KTH-TIPS2

The KTH-TIPS (Textures under varying Illumination, Pose and Scale) image database was created to extend the CUReT database in two directions, by providing variations in scale as well as pose and illumination, and by imaging other samples of a subset of its materials in different settings. The KTH-TIPS2 databases took this a step further by imaging 4 different samples of 11 materials, each under varying pose, illumination and scale. For more information about the databases, please view the documentation. The databases may be downloaded here.

Kuzushiji-Kanji

Kuzushiji-Kanji is an imbalanced dataset of total 3832 Kanji characters (64x64 grayscale, 140,426 images), ranging from 1,766 examples to only a single example per class. Kuzushiji is a Japanese cursive writing style. Source: Deep Learning for Classical Japanese Literature Image Source: Kuzushiji-MNIST

LIMUC

The LIMUC dataset is the largest publicly available labeled ulcerative colitis dataset that compromises 11276 images from 564 patients and 1043 colonoscopy procedures. Three experienced gastroenterologists were involved in the annotation process, and all images are labeled according to the Mayo endoscopic score (MES).

MIMIC-CXR-LT

MIMIC-CXR-LT. We construct a single-label, long-tailed version of MIMIC-CXR in a similar manner. MIMIC-CXR is a multi-label classification dataset with over 200,000 chest X-rays labeled with 13 pathologies and a “No Findings” class. The resulting MIMIC-CXR-LT dataset contains 19 classes, of which 10 are head classes, 6 are medium classes, and 3 are tail classes. MIMIC-CXR-LT contains 111,792 images labeled with one of 18 diseases, with 87,493 training images and 23,550 test set images. The validation and balanced test sets contain 15 and 30 images per class, respectively.

MNIST Large Scale dataset

The MNIST Large Scale dataset is based on the classic MNIST dataset, but contains large scale variations up to a factor of 16. The motivation behind creating this dataset was to enable testing the ability of different algorithms to learn in the presence of large scale variability and specifically the ability to generalise to new scales not present in the training set over wide scale ranges. The dataset contains training data for each one of the relative size factors 1, 2 and 4 relative to the original MNIST dataset and testing data for relative scaling factors between 1/2 and 8, with a ratio of $\sqrt[4]{2}$ between adjacent scales.

MUAD

The MUAD dataset (Multiple Uncertainties for Autonomous Driving), consisting of 10,413 realistic synthetic images with diverse adverse weather conditions (night, fog, rain, snow), out-of-distribution objects, and annotations for semantic segmentation, depth estimation, object, and instance detection. Predictive uncertainty estimation is essential for the safe deployment of Deep Neural Networks in real-world autonomous systems and MUAD allows to a better assess the impact of different sources of uncertainty on model performance.

Oracle-MNIST

We introduce the Oracle-MNIST dataset, comprising of 2828 grayscale images of 30,222 ancient characters from 10 categories, for benchmarking pattern classification, with particular challenges on image noise and distortion. The training set totally consists of 27,222 images, and the test set contains 300 images per class. Oracle-MNIST shares the same data format with the original MNIST dataset, allowing for direct compatibility with all existing classifiers and systems, but it constitutes a more challenging classification task than MNIST. The images of ancient characters suffer from 1) extremely serious and unique noises caused by three-thousand years of burial and aging and 2) dramatically variant writing styles by ancient Chinese, which all make them realistic for machine learning research. The dataset is freely available at https://github.com/wm-bupt/oracle-mnist.

WHU-Hi

WHU-Hi dataset (Wuhan UAV-borne hyperspectral image) is collected and shared by the RSIDEA research group of Wuhan University, and it could serve as a benchmark dataset for precise crop classification and hyperspectral image classification studies. The WHU-Hi dataset contains three individual UAV-borne hyperspectral datasets: WHU-Hi-LongKou, WHU-Hi-HanChuan, and WHU-Hi-HongHu. All the datasets were acquired in farming areas with various crop types in Hubei province, China, via a Headwall Nano-Hyperspec sensor mounted on a UAV platform. Compared with spaceborne and airborne hyperspectral platforms, unmanned aerial vehicle (UAV)-borne hyperspectral systems can acquire hyperspectral imagery with a high spatial resolution (which we refer to here as H2 imagery). The research was published in Remote Sensing of Environment. Source: http://rsidea.whu.edu.cn/resource_WHUHi_sharing.htm Image Source: http://rsidea.whu.edu.cn/resource_WHUHi_sharing.htm

ACL-Fig

ACL-Fig is a large-scale automatically annotated corpus consisting of 112,052 scientific figures extracted from 56K research papers in the ACL Anthology. The ACL-Fig-pilot dataset contains 1,671 manually labeled scientific figures belonging to 19 categories. Source: ACL-Fig: A Dataset for Scientific Figure Classification

Aircraft Context Dataset

The Aircraft Context Dataset, a composition of two inter-compatible large-scale and versatile image datasets focusing on manned aircraft and UAVs, is intended for training and evaluating classification, detection and segmentation models in aerial domains. Additionally, a set of relevant meta-parameters can be used to quantify dataset variability as well as the impact of environmental conditions on model performance.

Animals-10

It contains about 28K medium quality animal images belonging to 10 categories: dog, cat, horse, spyder, butterfly, chicken, sheep, cow, squirrel, and elephant. All the images have been collected from "google images" and have been checked by humans. There is some erroneous data to simulate real conditions (eg. images taken by users of your app). The main directory is divided into folders, one for each category. The image count for each category varies from 2K to 5 K units.

Atlas

Atlas is a dataset for e-commerce clothing product categorization. The Atlas dataset consists of a high-quality product taxonomy dataset focusing on clothing products which contain 186,150 images under clothing category with 3 levels and 52 leaf nodes in the taxonomy. Source: https://github.com/vumaasha/atlas Image Source: Umaashankaret et al

BarkNet 1.0

23,000 cropped images of tree bark, for 23 species of trees around Quebec City, Canada. The images were captured at a distance between 20-60 cm away from the trunk. Labels include: individual tree ID, its species, and its DBH (diameter at breast height). Pictures were taken with four different devices: Nexus 5, Samsung Galaxy S5, Samsung Galaxy S7, and a Panasonic Lumix DMC-TS5 camera. The dataset is sufficiently large to train a Deep network such as ResNet for species recognition.

Bengali.AI Handwritten Graphemes

This dataset contains images of individual hand-written Bengali characters. Bengali characters (graphemes) are written by combining three components: a grapheme_root, vowel_diacritic, and consonant_diacritic. Your challenge is to classify the components of the grapheme in each image. There are roughly 10,000 possible graphemes, of which roughly 1,000 are represented in the training set. The test set includes some graphemes that do not exist in the train but has no new grapheme components. It takes a lot of volunteers filling out sheets like this to generate a useful amount of real data; focusing the problem on the grapheme components rather than on recognizing whole graphemes should make it possible to assemble a Bengali OCR system without handwriting samples for all 10,000 graphemes.

DEIC Benchmark

DEIC is a benchmark for measuring the data efficiency of models in the context of image classification. It is composed of 6 datasets that contain a small number of training samples per class (i.e., 30 < x < 80). It covers multiple image domains (i.e., natural images, fine-grained recognition, medical images, remote sensing, handwriting recognition) and data types (i.e., RGB, grayscale, multi-spectral).

Diffusion Deepfake

Human face Deepfake dataset sampled from large datasets - High Quality Dataset - Diverse Dataset - Challenging Dataset - Large Dataset - Text prompts

Dirty-MNIST

DirtyMNIST is a concatenation of MNIST + AmbiguousMNIST, with 60k samples each in the training set. AmbiguousMNIST contains additional ambiguous digits with varying ambiguity. The AmbiguousMNIST test set contains 60k ambiguous samples as well. ## Additional Guidance 1. DirtyMNIST is a concatenation of MNIST + AmbiguousMNIST, with 60k samples each in the training set. 2. The current AmbiguousMNIST contains 6k unique samples with 10 labels each. This multi-label dataset gets flattened to 60k samples. The assumption is that ambiguous samples have multiple "valid" labels as they are ambiguous. MNIST samples are intentionally undersampled (in comparison), which benefits AL acquisition functions that can select unambiguous samples. 3. Pick your initial training samples (for warm starting Active Learning) from the MNIST half of DirtyMNIST to avoid starting training with potentially very ambiguous samples, which might add a lot of variance to your experiments. 4. Make sure to pick your validation set from the MNIST half as well, for the same reason as above. 5. Make sure that your batch acquisition size is >= 10 (probably) given that there are 10 multi-labels per samples in Ambiguous-MNIST. 6. By default, Gaussian noise with stddev 0.05 is added to each sample to prevent acquisition functions (in Active Learning) from cheating by disgarding "duplicates". 7. If you want to split Ambiguous-MNIST into subsets (or Dirty-MNIST within the second ambiguous half), make sure to split by multiples of 10 to avoid splits within a flattened multi-label sample.

FireRisk

In this work, we propose a novel remote sensing dataset, FireRisk, consisting of 7 fire risk classes with a total of 91 872 labelled images for fire risk assessment. This remote sensing dataset is labelled with the fire risk classes supplied by the Wildfire Hazard Potential (WHP) raster dataset, and remote sensing images are collected using the National Agriculture Imagery Program (NAIP), a high-resolution remote sensing imagery program. On FireRisk, we present benchmark performance for supervised and self-supervised representations, with Masked Autoencoders (MAE) pre-trained on ImageNet1k achieving the highest classification accuracy, 65.29%.

Geoclidean-Constraints

Geoclidean-Constraints dataset consists of 20 concepts and 40 tasks, created from permutations of line and circle construction rules with various constraints describing the relationship between objects. This dataset focuses on explicit constraints between geometric objects. We denote the objects as the following—lines as L, circles as C, and triangles (constructed from three lines) as T. Source: Geoclidean: Few-Shot Generalization in Euclidean Geometry Image Source: https://arxiv.org/pdf/2211.16663v1.pdf

HSI-Drive v2.0

HSI-Drive is the hyperspectral image (HSI) dataset created by the Digital Electronics Design Group (GDED) of the University of the Basque Country (UPV/EHU). This database is intended to contribute to the research into the use of hyperspectral imaging for the development of advanced driver assistance systems (ADAS) and autonomous driving systems (ADS). The dataset contains a diverse set of images recorded with a small-size 25-band VNIR snapshot camera mounted on a moving automobile. The recordings have been made in different seasons of the year, at different day times, under different weather conditions and on different types of roads. The dataset contains images and videos classified and tagged accordingly to provide rich and diverse data.

HyKo2-VIS

We present datasets containing urban traffic and rural road scenes recorded using hyperspectral snap-shot sensors mounted on a moving car. The novel hyperspectral cameras used can capture whole spectral cubes at up to 15 Hz. This emerging new sensor modality enables hyperspectral scene analysis for autonomous driving tasks. Up to the best of the author’s knowledge no such dataset has been published so far. The datasets contain synchronized 3-D laser, spectrometer and hyperspectral data. Dense ground truth annotations are provided as semantic labels, material and traversability. The hyperspectral data ranges from visible to near infrared wavelengths. We explain our recoding platform and method, the associated data format along with a code library for easy data consumption. The datasets are publicly available for download.

Image and Video Advertisements

The Image and Video Advertisements collection consists of an image dataset of 64,832 image ads, and a video dataset of 3,477 ads. The data contains rich annotations encompassing the topic and sentiment of the ads, questions and answers describing what actions the viewer is prompted to take and the reasoning that the ad presents to persuade the viewer ("What should I do according to this ad, and why should I do it? "), and symbolic references ads make (e.g. a dove symbolizes peace). Source: Automatic Understanding of Image and Video Advertisements

LSA16

This database contains images of 16 handshapes of the Argentinian Sign Language (LSA), each performed 5 times by 10 different subjects, for a total of 800 images. The subjects wore color hand gloves and dark clothes. Recording The dataset was recorded in an indoors environment, with artificial lightning. Subjects wore dark clothes and performed the handshapes standing, with a white wall as a background. To simplify the problem of hand segmentation, subjects wore fluorescent-colored gloves. These substantially simplify the problem of recognizing the position of the hand and performing its segmentation, and remove all issues associated to skin color variations, while fully retaining the difficulty of recognizing the handshape. The subjects performed the same handshape with both hands. Each handshape was executed imposing few constraints on the subjects to increase diversity and realism in the database. All subjects were non-signers and right-handed, were taught how to perform the handshape during the shooting session by showing them an image of the handshape as performed by one of the authors, and practiced each handshape a few times before recording. We employed a generic webcam for the recording, with a resolution of 640 by 480.

OFDIW

OnFocus Detection In the Wild (OFDIW) is an onfocus detection dataset. It consists of 20,623 images in unconstrained capture conditions (thus called "in the wild'') and contains individuals with diverse emotions, ages, facial characteristics, and rich interactions with surrounding objects and background scenes. The images are collected from the LFW dataset and the Oxford-IIIT Pet dataset. Onfocus detection aims at identifying whether the focus of the individual captured by a camera is on the camera or not.

RGB Arabic Alphabets Sign Language Dataset

This paper introduces the RGB Arabic Alphabet Sign Language (AASL) dataset. AASL comprises 7,856 raw and fully labeled RGB images of the Arabic sign language alphabets, which to our best knowledge is the first publicly available RGB dataset. The dataset is aimed to help those interested in developing real-life Arabic sign language classification models. AASL was collected from more than 200 participants and with different settings such as lighting, background, image orientation, image size, and image resolution. Experts in the field supervised, validated and filtered the collected images to ensure a high-quality dataset. AASL is made available to the public on Kaggle.

RP2K

A new large-scale retail product dataset for fine-grained image classification. Unlike previous datasets focusing on relatively few products, more than 500,000 images of retail products on shelves were collected, belonging to 2000 different products. The dataset aims to advance the research in retail object recognition, which has massive applications such as automatic shelf auditing and image-based product information retrieval. Source: RP2K: A Large-Scale Retail Product Dataset for Fine-Grained Image Classification

S-VED

The Sacrobosco Visual Elements Dataset (S-VED) is derived from 359 Sphaera editions, centered on the Tractatus de sphaera by Johannes de Sacrobosco (—1256) and printed between 1472 and 1650. The Sphaera editions were primarily used to teach geocentric astronomy to university students across Europe. Their visual elements, therefore, played an essential role in visualizing the ideas, messages, and concepts that the texts transmitted. As a precondition for studying the relation between text and visual elements, a time-consuming image labelling process was conducted as part of “The Sphere” project (https://sphaera.mpiwg-berlin.mpg.de) in order to extract and label the visual elements from the 76,000 pages of the corpus. This process resulted in the creation of the Extended Sacrobosco Visual Elements Dataset (S-VED𝑋) of which S-VED is a subset of. Due to copyright reasons only S-VED is made publicly available. S-VED consists of 4000 pages of which 2040 contain a total of 2927 visual elements. The visual elements are defined by bounding boxes and labels within a CSV file. For more information on the Sphaera corpus, feel free to check the project’s database at http://db.sphaera.mpiwg-berlin.mpg.de/.

SIDD-Image

This is the first image-based network intrusion detection dataset. This large-scale dataset included network traffic protocol communication-based images from 15 different observation locations of different countries in Asia. This dataset is used to identify two different types of anomalies from benign network traffic. Each image with a size of 48 × 48 contains multi-protocol communications within 128 seconds. The SIDD dataset can be to applied to a broad range of tasks such as machine learning-based network intrusion detection, non-iid federated learning, and so forth.

Stream-51

A new dataset for streaming classification consisting of temporally correlated images from 51 distinct object categories and additional evaluation classes outside of the training distribution to test novelty recognition. Source: Stream-51: Streaming Classification and Novelty Detection from Videos

AdvNet

AdvNet is a dataset of traffic signs images. Specifically, it includes adversarial traffic sign images (i.e., pictures of traffic signs with stickers on their surface) that can fool state-of-the-art neural network-based perception systems and clean traffic sign images without any stickers on them. If you use AdvNet, please cite the following paper: Y. Kantaros, T. Carpenter, K. Sridhar, I. Lee, J. Weimer: `Real-Time Detectors for Adversarial Digital and Physical Inputs to Perception Systems', 12th ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS), 2021

ASIRRA

Web services are often protected with a challenge that's supposed to be easy for people to solve, but difficult for computers. Such a challenge is often called a CAPTCHA (Completely Automated Public Turing test to tell Computers and Humans Apart) or HIP (Human Interactive Proof). HIPs are used for many purposes, such as to reduce email and blog spam and prevent brute-force attacks on web site passwords. Asirra (Animal Species Image Recognition for Restricting Access) is a HIP that works by asking users to identify photographs of cats and dogs. This task is difficult for computers, but studies have shown that people can accomplish it quickly and accurately. Many even think it's fun! Here is an example of the Asirra interface: Asirra is unique because of its partnership with Petfinder.com, the world's largest site devoted to finding homes for homeless pets. They've provided Microsoft Research with over three million images of cats and dogs, manually classified by people at thousands of animal shelters across the United States. Kaggle is fortunate to offer a subset of this data for fun and research.

출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.