Datasets

9,767 machine learning datasets
🔔   Share your dataset with the ML community!
Filter by Modality
Images 25 3D 20 RGB-D 9 Videos 9 3d meshes 5 Point cloud 3 Stereo 2 Biology 1 Cad 1 Hyperspectral images 1 Medical 1 Texts 1 6D 0 Actions 0 Audio 0 Biomedical 0 Dialog 0 EEG 0 Environment 0 Financial 0 Graphs 0 Interactive 0 LiDAR 0 Lyrics 0 MRI 0 Midi 0 Music 0 PSG 0 Parallel 0 Physics 0 RGB Video 0 Ranking 0 Replay data 0 Speech 0 Tables 0 Tabular 0 Time series 0 Tracking 0 fMRI 0
Filter by Task (clear)
3D Reconstruction Question Answering 360 Semantic Segmentation 311 Object Detection 261 Image Classification 240 Language Modelling 157 Text Generation 150 Text Classification 148 Classification 125 Named Entity Recognition (NER) 122 Pose Estimation 113 Visual Question Answering (VQA) 112 Action Recognition 105 Anomaly Detection 95 Reading Comprehension 95 Sentiment Analysis 93 Domain Adaptation 88 Speech Recognition 87 Text Summarization 87 Information Retrieval 82 Instance Segmentation 82 Natural Language Inference 77 Machine Translation 76 Image Retrieval 75 Relation Extraction 74 Natural Language Understanding 71 Depth Estimation 70 Node Classification 69 Image Generation 67 Autonomous Driving 66 Image Captioning 65 Data Augmentation 63 Link Prediction 63 Face Recognition 61 Object Tracking 61 2D Object Detection 57 2D Semantic Segmentation 57 Person Re-Identification 57 Multi-Task Learning 55 Recommendation Systems 54 Common Sense Reasoning 52 Word Embeddings 52 3D Object Detection 48 Abstractive Text Summarization 48 3D Human Pose Estimation 46 Graph Classification 46 Emotion Recognition 45 Code Generation 43 Coreference Resolution 43 Medical Image Segmentation 43 Scene Understanding 43 Object Recognition 42 Optical Character Recognition (OCR) 42 Semantic Parsing 42 Temporal Action Localization 42 Video Understanding 42 Few-Shot Learning 41 Knowledge Graphs 41 Machine Reading Comprehension 41 Self-Supervised Learning 41 Visual Reasoning 41 Hate Speech Detection 39 Image Super-Resolution 39 Decision Making 38 Entity Linking 38 Misinformation 38 Multi-Object Tracking 37 Face Detection 36 Fine-Grained Image Classification 36 Audio Classification 34 Trajectory Prediction 34 Action Detection 33 Novel View Synthesis 33 Optical Flow Estimation 33 Panoptic Segmentation 32 Video Captioning 32 Video Question Answering 32 3D Semantic Segmentation 31 Dialogue Generation 31 Metric Learning 31 Stance Detection 31 Unsupervised Domain Adaptation 31 Video Retrieval 31 Action Classification 30 3D Pose Estimation 29 Activity Recognition 29 Continual Learning 29 Facial Expression Recognition (FER) 29 Image-to-Image Translation 29 Skeleton Based Action Recognition 29 Zero-Shot Learning 29 Document Summarization 28 Time Series Forecasting 28 Token Classification 28 Autonomous Vehicles 27 Emotion Classification 27 Multi-Label Classification 27 Neural Architecture Search 27 Scene Text Recognition 27 Monocular Depth Estimation 26 Open-Domain Question Answering 26 Part-Of-Speech Tagging 26 Sequence-to-sequence Language Modeling 26 Visual Object Tracking 26 Visual Tracking 26 Domain Generalization 25 Fake News Detection 25 Question Generation 25 Retrieval 25 Drug Discovery 24 Music Generation 24 NER 24 Video Prediction 24 Automatic Speech Recognition (ASR) 23 Few-Shot Image Classification 23 Music Information Retrieval 23 Object Counting 23 Unsupervised Anomaly Detection 23 2D Human Pose Estimation 22 Data-to-Text Generation 22 Hand Pose Estimation 22 Human-Object Interaction Detection 22 Relation Classification 22 Face Verification 21 Low-Light Image Enhancement 21 Out-of-Distribution Detection 21 Visual Odometry 21 Cross-Modal Retrieval 20 Denoising 20 Fairness 20 Handwriting Recognition 20 Image Clustering 20 Image Segmentation 20 Intent Detection 20 Scene Classification 20 Slot Filling 20 Time Series Analysis 20 Age Estimation 19 Crowd Counting 19 Meta-Learning 19 Molecular Property Prediction 19 Speech Synthesis 19 Super-Resolution 19 Task-Oriented Dialogue Systems 19 Text Simplification 19 Visual Localization 19 Visual Place Recognition 19 Visual Question Answering 19 Aspect-Based Sentiment Analysis (ABSA) 18 Cell Segmentation 18 DeepFake Detection 18 Imitation Learning 18 Language Identification 18 Quantization 18 Sign Language Recognition 18 Simultaneous Localization and Mapping 18 Sound Event Detection 18 Speech Emotion Recognition 18 Speech Enhancement 18 Text-to-Image Generation 18 Traffic Prediction 18 6D Pose Estimation 17 Action Recognition In Videos 17 Animal Pose Estimation 17 Binarization 17 Face Alignment 17 Face Anti-Spoofing 17 Graph Clustering 17 Graph Regression 17 Motor Imagery Decoding (left-hand vs right-hand) 17 Multimodal Deep Learning 17 Paraphrase Identification 17 SMAC+ 17 Semantic Textual Similarity 17 Stereo Matching 17 Text-To-Speech Synthesis 17 Video Classification 17 3D Hand Pose Estimation 16 Action Segmentation 16 Cross-Lingual Transfer 16 Image Dehazing 16 Image Inpainting 16 Knowledge Graph Completion 16 Long-tail Learning 16 Math Word Problem Solving 16 Object Localization 16 Paraphrase Generation 16 Prompt Engineering 16 RGB Salient Object Detection 16 Robotic Grasping 16 Salient Object Detection 16 Speech Separation 16 Style Transfer 16 Time Series Classification 16 Video Object Segmentation 16 Visual Navigation 16 Active Learning 15 Document Classification 15 Facial Landmark Detection 15 Gaze Estimation 15 Grammatical Error Correction 15 Image Denoising 15 Image Enhancement 15 Mathematical Reasoning 15 Medical Diagnosis 15 Multiple Object Tracking 15 Node Classification on Non-Homophilic (Heterophilic) Graphs 15 Reinforcement Learning (RL) 15 Scene Recognition 15 Scene Text Detection 15 Translation 15 3D Action Recognition 14 Change Detection 14 Computed Tomography (CT) 14 Deblurring 14 Dependency Parsing 14 Fact Verification 14 Instruction Following 14 Joint Entity and Relation Extraction 14 Lane Detection 14 Multi-Document Summarization 14 Node Clustering 14 Self-Driving Cars 14 Semantic Role Labeling 14 Spoken Language Understanding 14 Text-To-SQL 14 Trajectory Forecasting 14 Word Sense Disambiguation 14 Zero-shot Text Search 14 3D Human Reconstruction 13 3D Instance Segmentation 13 Audio Source Separation 13 Clustering Algorithms Evaluation 13 Density Estimation 13 Emotion Recognition in Conversation 13 Event Extraction 13 Hand Gesture Recognition 13 Intent Classification 13 License Plate Recognition 13 Motion Synthesis 13 Open Information Extraction 13 Pedestrian Detection 13 Robot Navigation 13 Sarcasm Detection 13 Semi-Supervised Video Object Segmentation 13 Sentence Classification 13 Transfer Learning 13 Video Generation 13 Activity Detection 12 Binary Classification 12 Disentanglement 12 Entity Disambiguation 12 Gesture Recognition 12 Human Detection 12 Image Quality Assessment 12 Learning with noisy labels 12 Lesion Segmentation 12 Semi-Supervised Image Classification 12 Single-View 3D Reconstruction 12 Text Retrieval 12 Topic Models 12 Video Quality Assessment 12 Video Summarization 12 Video Super-Resolution 12 Vision and Language Navigation 12 Weakly Supervised Object Detection 12 Weather Forecasting 12 3D Classification 11 3D Face Reconstruction 11 3D Object Tracking 11 Automatic Speech Recognition 11 Community Detection 11 Conversational Response Selection 11 Dialogue State Tracking 11 Entity Resolution 11 Entity Typing 11 Image Registration 11 Image Restoration 11 Keypoint Detection 11 Logical Reasoning 11 Motion Forecasting 11 Multi-Label Text Classification 11 Multi-class Classification 11 Network Intrusion Detection 11 Open-Domain Dialog 11 Real-Time Semantic Segmentation 11 Relational Reasoning 11 Saliency Detection 11 Semantic Similarity 11 Semi-Supervised Semantic Segmentation 11 Sign Language Translation 11 Stochastic Optimization 11 Surface Normals Estimation 11 Video Frame Interpolation 11 Zero-Shot Video Question Answer 11 motion prediction 11 Acoustic Scene Classification 10 Answer Selection 10 Boundary Detection 10 COVID-19 Diagnosis 10 Code Search 10 Column Type Annotation 10 Contrastive Learning 10 Depth Completion 10 Dimensionality Reduction 10 Facial Attribute Classification 10 Federated Learning 10 Graph Embedding 10 Image Compression 10 Imputation 10 Named Entity Recognition 10 Nested Named Entity Recognition 10 Object Detection In Indoor Scenes 10 Outlier Detection 10 Referring Expression Segmentation 10 Sentence Embeddings 10 Small Object Detection 10 Table annotation 10 Time Series Prediction 10 Unsupervised Object Segmentation 10 Video Object Tracking 10 Virtual Try-on 10 3D Medical Imaging Segmentation 9 Abusive Language 9 Adversarial Robustness 9 Automatic Post-Editing 9 Citation Recommendation 9 Code Completion 9 Continuous Control 9 Conversational Question Answering 9 Cross-Lingual NER 9 Dialogue Understanding 9 Edge Detection 9 Few-Shot Audio Classification 9 Fine-Grained Image Recognition 9 Fraud Detection 9 Gait Recognition 9 Graph Matching 9 Handwriting generation 9 Head Pose Estimation 9 Incremental Learning 9 License Plate Detection 9 Material Recognition 9 Medical Image Classification 9 News Classification 9 Opinion Mining 9 Point Cloud Registration 9 Sentiment Classification 9 Speaker Diarization 9 Unsupervised Person Re-Identification 9 Video Inpainting 9 Video Object Detection 9 Video Recognition 9 Visual Dialog 9 2D Pose Estimation 8 3D Depth Estimation 8 Ad-hoc video search 8 Adversarial Attack 8 Audio Generation 8 Audio Tagging 8 Automated Theorem Proving 8 Causal Inference 8 Chart Question Answering 8 Code Translation 8 Color Image Denoising 8 Defect Detection 8 Discourse Parsing 8 Document Layout Analysis 8 ECG Classification 8 Entity Alignment 8 Event-based vision 8 Explanation Generation 8 Face Swapping 8 Fact Checking 8 Font Recognition 8 General Classification 8 Generalizable Person Re-identification 8 Generalized Zero-Shot Learning 8 Handwritten Text Recognition 8 Human Activity Recognition 8 Key Information Extraction 8 Medical Image Registration 8 Multi-Label Image Classification 8 Multi-Label Learning 8 Multi-agent Reinforcement Learning 8 Music Transcription 8 Online Multi-Object Tracking 8 Person Search 8 Pose Tracking 8 Referring Expression Comprehension 8 Scene Generation 8 Segmentation 8 Temporal Tagging 8 Text-to-Video Generation 8 Time Series 8 Tumor Segmentation 8 Unsupervised Video Object Segmentation 8 Vehicle Re-Identification 8 Video Segmentation 8 regression 8 3D Human Shape Estimation 7 3D Multi-Object Tracking 7 3D Object Recognition 7 3D Shape Reconstruction 7 Action Anticipation 7 Action Quality Assessment 7 AutoML 7 Bias Detection 7 Blind Super-Resolution 7 Change Point Detection 7 Chatbot 7 Chinese Reading Comprehension 7 Code Summarization 7 Colorectal Polyps Characterization 7 Conditional Image Generation 7 Data Integration 7 Dense Video Captioning 7 Dialogue Act Classification 7 Disaster Response 7 Dynamic Link Prediction 7 Electroencephalogram (EEG) 7 Event Detection 7 Explainable artificial intelligence 7 Food Recognition 7 Formation Energy 7 Human Part Segmentation 7 KG-to-Text Generation 7 Keyword Spotting 7 Knowledge Base Question Answering 7 Multimodal Activity Recognition 7 Multimodal Emotion Recognition 7 Multiple-choice 7 Multiview Detection 7 Passage Retrieval 7 Physical Simulations 7 Product Recommendation 7 Program Repair 7 Remote Sensing Image Classification 7 Robust Object Detection 7 Single Image Dehazing 7 Skin Lesion Classification 7 Source Code Summarization 7 Stance Classification 7 Table Detection 7 Text Segmentation 7 Traffic Sign Recognition 7 Translation deu-eng 7 Translation eng-deu 7 Unsupervised Semantic Segmentation with Language-image Pre-training 7 Video Anomaly Detection 7 Video Grounding 7 Video Semantic Segmentation 7 Zero-Shot Transfer Image Classification 7 3D Absolute Human Pose Estimation 6 3D Face Animation 6 3D Object Classification 6 3D Object Reconstruction 6 3D Part Segmentation 6 3D Point Cloud Classification 6 6D Pose Estimation using RGB 6 6D Pose Estimation using RGBD 6 Argument Mining 6 Atari Games 6 Audio-Visual Speech Recognition 6 Band Gap 6 Bayesian Inference 6 Body Detection 6 Breast Cancer Detection 6 Camouflaged Object Segmentation 6 Click-Through Rate Prediction 6 Code Classification 6 Code Repair 6 Colorization 6 Compositional Zero-Shot Learning 6 Constituency Parsing 6 Cross-Domain Few-Shot 6 Cross-Lingual Question Answering 6 Dialogue Evaluation 6 Dictionary Learning 6 Egocentric Pose Estimation 6 Emotion Recognition in Context 6 Environmental Sound Classification 6 Extreme Summarization 6 Facial Emotion Recognition 6 Few-Shot Object Detection 6 Gesture Generation 6 Graph Learning 6 Heart rate estimation 6 Hierarchical Multi-label Classification 6 Human Interaction Recognition 6 Hyperspectral Image Classification 6 Lesion Classification 6 Medical Visual Question Answering 6 Motion Estimation 6 Multi-Person Pose Estimation 6 Multiple Instance Learning 6 Multivariate Time Series Forecasting 6 Multivariate Time Series Imputation 6 Person Identification 6 Real-Time Object Detection 6
Filter by Language
English 9 Abkhazian 0 Achinese 0 Adyghe 0 Afar 0 Afrikaans 0 Akan 0 Akkadian 0 Akuntsu 0 Albanian 0 American Sign Language 0 Amharic 0 Ancient Greek 0 Ancient Hebrew 0 Apurinã 0 Arabic 0 Aragonese 0 Argentine Sign Language 0 Armenian 0 Arpitan 0 Assamese 0 Assyrian Neo-Aramaic 0 Asturian 0 Avaric 0 Aymara 0 Azerbaijani 0 Bambara 0 Bangala 0 Bangladeshi Sign Language 0 Banjar 0 Bashkir 0 Basque 0 Bavarian 0 Belarusian 0 Bemba (Zambia) 0 Bengali 0 Bhojpuri 0 Bishnupriya 0 Bislama 0 Bodo (India) 0 Bosnian 0 Breton 0 Buginese 0 Bulgarian 0 Burmese 0 Catalan 0 Cebuano 0 Central Bikol 0 Central Khmer 0 Central Kurdish 0 Central Pashto 0 Chamorro 0 Chavacano 0 Chechen 0 Cherokee 0 Cheyenne 0 Chinese 0 Choctaw 0 Chukot 0 Church Slavic 0 Chuvash 0 Congo Swahili 0 Coptic 0 Cornish 0 Corsican 0 Cree 0 Creek 0 Crimean Tatar 0 Croatian 0 Czech 0 Danish 0 Dhivehi 0 Dimli (individual language) 0 Dogri (individual language) 0 Dogri (macrolanguage) 0 Dutch 0 Dzongkha 0 Eastern Mari 0 Egyptian Arabic 0 Erzya 0 Esperanto 0 Estonian 0 Ewe 0 Extremaduran 0 Faroese 0 Fiji Hindi 0 Fijian 0 Filipino 0 Finnish 0 Fon 0 French 0 Friulian 0 Fulah 0 Gagauz 0 Galician 0 Gan Chinese 0 Ganda 0 Geez 0 Georgian 0 German 0 German Sign Language 0 Gilaki 0 Goan Konkani 0 Gothic 0 Greek 0 Greek Sign Language 0 Guarani 0 Gujarati 0 Gulf Arabic 0 Haitian 0 Hakha Chin 0 Hakka Chinese 0 Halh Mongolian 0 Hausa 0 Hawaiian 0 Hebrew 0 Herero 0 Hindi 0 Hiri Motu 0 Hungarian 0 Icelandic 0 Ido 0 Igbo 0 Iloko 0 Indonesian 0 Interlingua (International Auxiliary Language Association) 0 Interlingue 0 Inuktitut 0 Inupiaq 0 Iranian Persian 0 Irish 0 Italian 0 Jamaican Creole English 0 Japanese 0 Javanese 0 Jejueo 0 Kabardian 0 Kabuverdianu 0 Kabyle 0 Kachin 0 Kalaallisut 0 Kalmyk 0 Kannada 0 Kanuri 0 Kara-Kalpak 0 Karachay-Balkar 0 Karelian 0 Kashmiri 0 Kashubian 0 Kazakh 0 Khunsari 0 Kikuyu 0 Kinyarwanda 0 Kirghiz 0 Komi 0 Komi-Permyak 0 Komi-Zyrian 0 Kongo 0 Korean 0 Krio 0 Kuanyama 0 Kurdish 0 Kölsch 0 Ladino 0 Lak 0 Lao 0 Latgalian 0 Latin 0 Latvian 0 Lezghian 0 Ligurian 0 Limburgan 0 Lingala 0 Literary Chinese 0 Lithuanian 0 Livvi 0 Lojban 0 Lombard 0 Low German 0 Lower Sorbian 0 Lozi 0 Lunda 0 Luo (Cameroon) 0 Luo (Kenya and Tanzania) 0 Lushai 0 Luxembourgish 0 Macedonian 0 Maithili 0 Malagasy 0 Malay (individual language) 0 Malay (macrolanguage) 0 Malayalam 0 Maltese 0 Mandarin Chinese 0 Manipuri 0 Manx 0 Maori 0 Marathi 0 Marshallese 0 Mazanderani 0 Mbyá Guaraní 0 Mesopotamian Arabic 0 Min Dong Chinese 0 Minangkabau 0 Mingrelian 0 Mirandese 0 Modern Greek 0 Modern Greek (1453-) 0 Moksha 0 Mongolian 0 Moroccan Arabic 0 Multilingual 0 Mundurukú 0 Najdi Arabic 0 Narom 0 Nauru 0 Navajo 0 Naxi 0 Nayini 0 Ndonga 0 Neapolitan 0 Nepali (individual language) 0 Nepali (macrolanguage) 0 Newari 0 Nigerian Fulfulde 0 Nigerian Pidgin 0 North Azerbaijani 0 North Levantine Arabic 0 Northern Frisian 0 Northern Huishui Hmong 0 Northern Kurdish 0 Northern Luri 0 Northern Sami 0 Northern Uzbek 0 Norwegian 0 Norwegian Bokmål 0 Norwegian Nynorsk 0 Novial 0 Nyanja 0 Occitan (post 1500) 0 Odia 0 Official Aramaic (700-300 BCE) 0 Old English (ca. 450-1100) 0 Old French 0 Old Russian 0 Old Turkish 0 Oriya (macrolanguage) 0 Oromo 0 Ossetian 0 Pali 0 Pampanga 0 Pangasinan 0 Papiamento 0 Pedi 0 Pennsylvania German 0 Persian 0 Pfaelzisch 0 Picard 0 Piemontese 0 Pitcairn-Norfolk 0 Plateau Malagasy 0 Polish 0 Pontic 0 Portuguese 0 Portuguse 0 Punjabi 0 Pushto 0 Quechua 0 Rajasthani 0 Romanian 0 Romansh 0 Rundi 0 Russia Buriat 0 Russian 0 Rusyn 0 Saidi Arabic 0 Samoan 0 Sango 0 Sanskrit 0 Santali 0 Sardinian 0 Saterfriesisch 0 Scots 0 Scottish Gaelic 0 Serbian 0 Serbo-Croatian 0 Shan 0 Shona 0 Sichuan Yi 0 Sicilian 0 Silesian 0 Sindhi 0 Sinhala 0 Skolt Sami 0 Slovak 0 Slovenian 0 Soi 0 Somali 0 South Azerbaijani 0 South Levantine Arabic 0 Southern Pashto 0 Southern Sotho 0 Spanish 0 Sranan Tongo 0 Standard Arabic 0 Standard Latvian 0 Sundanese 0 Swahili 0 Swahili (macrolanguage) 0 Swati 0 Swedish 0 Swedish Sign Language 0 Swiss German 0 Swiss-German Sign Language 0 Tagalog 0 Tahitian 0 Tai 0 Tajik 0 Tamil 0 Tatar 0 Telugu 0 Tetum 0 Thai 0 Tibetan 0 Tigrinya 0 Tok Pisin 0 Tonga (Tonga Islands) 0 Tonga (Zambia) 0 Tosk Albanian 0 Tsonga 0 Tswana 0 Tulu 0 Tumbuka 0 Tunisian Arabic 0 Tupinambá 0 Turkish 0 Turkish Sign Language 0 Turkmen 0 Tuvinian 0 Twi 0 Udmurt 0 Uighur 0 Ukrainian 0 Upper Sorbian 0 Urdu 0 Uzbek 0 Venda 0 Venetian 0 Veps 0 Vietnamese 0 Vlaams 0 Vlax Romani 0 Volapük 0 Votic 0 Walloon 0 Waray (Philippines) 0 Warlpiri 0 Welsh 0 West Central Oromo 0 Western Frisian 0 Western Mari 0 Western Panjabi 0 Wolof 0 Wu Chinese 0 Xhosa 0 Yakut 0 Yiddish 0 Yoruba 0 Yue Chinese 0 Zaza 0 Zeeuws 0 Zhuang 0 Zulu 0

54 dataset results for 3D Reconstruction

ShapeNet

ShapeNet is a large scale repository for 3D CAD models developed by researchers from Stanford University, Princeton University and the Toyota Technological Institute at Chicago, USA. The repository contains over 300M models with 220,000 classified into 3,135 classes arranged using WordNet hypernym-hyponym relationships. ShapeNet Parts subset contains 31,693 meshes categorised into 16 common object classes (i.e. table, chair, plane etc.). Each shapes ground truth contains 2-5 parts (with a total of 50 part classes).

1,689 PAPERS • 13 BENCHMARKS

ScanNet

ScanNet is an instance-level indoor RGB-D dataset that includes both 2D and 3D data. It is a collection of labeled voxels rather than points or objects. Up to now, ScanNet v2, the newest version of ScanNet, has collected 1513 annotated scans with an approximate 90% surface coverage. In the semantic segmentation task, this dataset is marked in 20 classes of annotated 3D voxelized objects.

1,240 PAPERS • 19 BENCHMARKS

DTU (DTU MVS dataset - 2014)

DTU MVS 2014 is a multi-view stereo dataset, which is an order of magnitude larger in number of scenes and with a significant increase in diversity. Specifically, it contains 80 scenes of large variability. Each scene consists of 49 or 64 accurate camera positions and reference structured light scans, all acquired by a 6-axis industrial robot.

219 PAPERS • 2 BENCHMARKS

300W (300 Faces-In-The-Wild)

The 300-W is a face dataset that consists of 300 Indoor and 300 Outdoor in-the-wild images. It covers a large variation of identity, expression, illumination conditions, pose, occlusion and face size. The images were downloaded from google.com by making queries such as “party”, “conference”, “protests”, “football” and “celebrities”. Compared to the rest of in-the-wild datasets, the 300-W database contains a larger percentage of partially-occluded images and covers more expressions than the common “neutral” or “smile”, such as “surprise” or “scream”. Images were annotated with the 68-point mark-up using a semi-automatic methodology. The images of the database were carefully selected so that they represent a characteristic sample of challenging but natural face instances under totally unconstrained conditions. Thus, methods that achieve accurate performance on the 300-W database can demonstrate the same accuracy in most realistic cases. Many images of the database contain more than one a

198 PAPERS • 9 BENCHMARKS

ShapeNetCore

ShapeNetCore is a subset of the full ShapeNet dataset with single clean 3D models and manually verified category and alignment annotations. It covers 55 common object categories with about 51,300 unique 3D models. The 12 object categories of PASCAL 3D+, a popular computer vision 3D benchmark dataset, are all covered by ShapeNetCore.

156 PAPERS • 1 BENCHMARK

MegaDepth

The MegaDepth dataset is a dataset for single-view depth prediction that includes 196 different locations reconstructed from COLMAP SfM/MVS.

115 PAPERS • NO BENCHMARKS YET

ABC Dataset

The ABC Dataset is a collection of one million Computer-Aided Design (CAD) models for research of geometric deep learning methods and applications. Each model is a collection of explicitly parametrized curves and surfaces, providing ground truth for differential quantities, patch segmentation, geometric feature detection, and shape reconstruction. Sampling the parametric descriptions of surfaces and curves allows generating data in different formats and resolutions, enabling fair comparisons for a wide range of geometric learning algorithms.

83 PAPERS • NO BENCHMARKS YET

ETH3D

ETHD is a multi-view stereo benchmark / 3D reconstruction benchmark that covers a variety of indoor and outdoor scenes. Ground truth geometry has been obtained using a high-precision laser scanner. A DSLR camera as well as a synchronized multi-camera rig with varying field-of-view was used to capture images.

79 PAPERS • 1 BENCHMARK

BlendedMVS

BlendedMVS is a novel large-scale dataset, to provide sufficient training ground truth for learning-based MVS. The dataset was created by applying a 3D reconstruction pipeline to recover high-quality textured meshes from images of well-selected scenes. Then, these mesh models were rendered to color images and depth maps.

74 PAPERS • NO BENCHMARKS YET

Scan2CAD

Scan2CAD is an alignment dataset based on 1506 ScanNet scans with 97607 annotated keypoints pairs between 14225 (3049 unique) CAD models from ShapeNet and their counterpart objects in the scans. The top 3 annotated model classes are chairs, tables and cabinets which arises due to the nature of indoor scenes in ScanNet. The number of objects aligned per scene ranges from 1 to 40 with an average of 9.3.

65 PAPERS • 1 BENCHMARK

Hypersim

For many fundamental scene understanding tasks, it is difficult or impossible to obtain per-pixel ground truth labels from real images. Hypersim is a photorealistic synthetic dataset for holistic indoor scene understanding. It contains 77,400 images of 461 indoor scenes with detailed per-pixel labels and corresponding ground truth geometry.

59 PAPERS • 1 BENCHMARK

ABO (Amazon Berkeley Objects)

ABO is a large-scale dataset designed for material prediction and multi-view retrieval experiments. The dataset contains Blender renderings of 30 viewpoints for each of the 7,953 3D objects, as well as camera intrinsics and extrinsic for each rendering.

46 PAPERS • NO BENCHMARKS YET

Tanks and Temples

We present a benchmark for image-based 3D reconstruction. The benchmark sequences were acquired outside the lab, in realistic conditions. Ground-truth data was captured using an industrial laser scanner. The benchmark includes both outdoor scenes and indoor environments. High-resolution video sequences are provided as input, supporting the development of novel pipelines that take advantage of video input to increase reconstruction fidelity. We report the performance of many image-based 3D reconstruction pipelines on the new benchmark. The results point to exciting challenges and opportunities for future work.

39 PAPERS • 2 BENCHMARKS

People Snapshot Dataset
People Snapshot Dataset

Enables detailed human body model reconstruction in clothing from a single monocular RGB video without requiring a pre scanned template or manually clicked points.

33 PAPERS • NO BENCHMARKS YET

Dynamic FAUST

Dynamic FAUST extends the FAUST dataset to dynamic 4D data. It consists of high-resolution 4D scans of human subjects in motion, captured at 60 fps.

28 PAPERS • 1 BENCHMARK

ApolloCar3D

ApolloCar3DT is a dataset that contains 5,277 driving images and over 60K car instances, where each car is fitted with an industry-grade 3D CAD model with absolute model size and semantically labelled keypoints. This dataset is above 20 times larger than PASCAL3D+ and KITTI, the current state-of-the-art.

17 PAPERS • 14 BENCHMARKS

Common Objects in 3D

Common Objects in 3D is a large-scale dataset with real multi-view images of object categories annotated with camera poses and ground truth 3D point clouds. The dataset contains a total of 1.5 million frames from nearly 19,000 videos capturing objects from 50 MS-COCO categories and, as such, it is significantly larger than alternatives both in terms of the number of categories and objects.

17 PAPERS • 1 BENCHMARK

SceneNet

SceneNet is a dataset of labelled synthetic indoor scenes. There are several labeled indoor scenes, including:

17 PAPERS • NO BENCHMARKS YET

DublinCity
DublinCity

A novel benchmark dataset that includes a manually annotated point cloud for over 260 million laser scanning points into 100'000 (approx.) assets from Dublin LiDAR point cloud [12] in 2015. Objects are labelled into 13 classes using hierarchical levels of detail from large (i.e., building, vegetation and ground) to refined (i.e., window, door and tree) elements.

13 PAPERS • NO BENCHMARKS YET

ARCTIC
ARCTIC (Articulated Objects in Free-form Hand Interaction)

ARCTIC is a dataset of free-form interactions of hands and articulated objects. ARCTIC has 1.2M images paired with accurate 3D meshes for both hands and for objects that move and deform over time. The dataset also provides hand-object contact information.

9 PAPERS • NO BENCHMARKS YET

ETH SfM (ETH Structure-from-Motion)

The ETH SfM (structure-from-motion) dataset is a dataset for 3D Reconstruction. The benchmark investigates how different methods perform in terms of building a 3D model from a set of available 2D images.

8 PAPERS • NO BENCHMARKS YET

Oxford-Affine

The Oxford-Affine dataset is a small dataset containing 8 scenes with sequence of 6 images per scene. The images in a sequence are related by homographies.

7 PAPERS • NO BENCHMARKS YET

3DPeople Dataset
3DPeople Dataset

A large-scale synthetic dataset with 2.5 Million photo-realistic images of 80 subjects performing 70 activities and wearing diverse outfits.

6 PAPERS • NO BENCHMARKS YET

MobileBrick
MobileBrick

Generate high-quality 3D ground-truth shapes for reconstruction evaluation is extremely challenging because even 3D scanners can only generate pseudo ground-truth shapes with artefacts. We propose a novel data capturing and 3D annotation pipeline to obtain precise 3D ground-truth shapes without relying on expensive 3D scanners. The key to creating the precise 3D ground-truth shapes is using LEGO models, which are made of LEGO bricks with known geometry. The MobileBrick dataset provides a unique opportunity for future research on high-quality 3D reconstruction thanks to two distinctive features: 1) A large number of RGBD sequences with precise 3D ground-truth annotations. 2) The RGBD images were captured using mobile devices so algorithms can be tested in a realistic setup for mobile AR applications.

6 PAPERS • NO BENCHMARKS YET

MonoPerfCap Dataset

MonoPerfCap is a benchmark dataset for human 3D performance capture from monocular video input consisting of around 40k frames, which covers a variety of different scenarios.

6 PAPERS • NO BENCHMARKS YET

A Large Dataset of Object Scans

A Large Dataset of Object Scans is a dataset of more than ten thousand 3D scans of real objects. To create the dataset, the authors recruited 70 operators, equipped them with consumer-grade mobile 3D scanning setups, and paid them to scan objects in their environments. The operators scanned objects of their choosing, outside the laboratory and without direct supervision by computer vision professionals. The result is a large and diverse collection of object scans: from shoes, mugs, and toys to grand pianos, construction vehicles, and large outdoor sculptures. The authors worked with an attorney to ensure that data acquisition did not violate privacy constraints. The acquired data was placed in the public domain and is available freely.

5 PAPERS • NO BENCHMARKS YET

HOD (Hand-held Object Dataset)

HOD is a dataset for 3D object reconstruction which contains 35 objects, divided into two subsets named Sculptures and Daily Objects. The Sculptures has five human sculptures with complex geometries and pure white textures. The Daily Objects consists of 30 daily objects with various shapes and appearances. All of the Sculptures and nine of the Daily Objects are paired with high-fidelity scanned meshes as ground truth geometries for evaluation.

4 PAPERS • NO BENCHMARKS YET

Houses3K

Houses3K is a dataset of 3000 textured 3D house models. Houses3K is divided into twelve batches, each containing 50 unique house geometries. For each batch, five different textures were applied forming the sets (A, B, C, D, E).

4 PAPERS • NO BENCHMARKS YET

CoP3D

CoP3D is a collection of crowd-sourced videos showing around 4,200 distinct pets. CoP2D is a large-scale datasets for benchmarking non-rigid 3D reconstruction "in the wild".

3 PAPERS • NO BENCHMARKS YET

Dynamic Replica

Dynamic Replica is a synthetic dataset of stereo videos featuring humans and animals in virtual environments. It is a benchmark for dynamic disparity/depth estimation and 3D reconstruction consisting of 145,200 stereo frames (524 videos).

3 PAPERS • NO BENCHMARKS YET

Parcel2D Real

Real-world dataset of ~400 images of cuboid-shaped parcels with full 2D and 3D annotations in the COCO format.

3 PAPERS • NO BENCHMARKS YET

ShapenetRender

ShapenetRenderer is an extension of the ShapeNet Core dataset which has more variation in camera angles. For each mesh model, the dataset provides 36 views with smaller variation and 36 views with larger variation. The resolution of the newly rendered images is 224x224 in contrast to the 137x137 original resolution. Additionally, each RGB image is paired with a depth image, a normal map and an albedo image.

3 PAPERS • NO BENCHMARKS YET

Tragic Talkers

Tragic Talkers is an audio-visual dataset consisting of excerpts from the "Romeo and Juliet" drama captured with microphone arrays and multiple co-located cameras for light-field video. Tragic Talkers provides ideal content for object-based media (OBM) production. It is designed to cover various conventional talking scenarios, such as monologues, two-people conversations, and interactions with considerable movement and occlusion, yielding 30 sequences captured from a total of 22 different points of view and two 16-element microphone arrays.

3 PAPERS • NO BENCHMARKS YET

ENRICH
ENRICH (Multi-purposE dataset for beNchmaRking In Computer vision and pHotogrammetry)

A new synthetic, multi-purpose dataset - called ENRICH - for testing photogrammetric and computer vision algorithms. Compared to existing datasets, ENRICH offers higher resolution images also rendered with different lighting conditions, camera orientation, scales, and field of view. Specifically, ENRICH is composed of three sub-datasets: ENRICH-Aerial, ENRICH-Square, and ENRICH-Statue, each exhibiting different characteristics. The proposed dataset is useful for several photogrammetry and computer vision-related tasks, such as the evaluation of hand-crafted and deep learning-based local features, effects of ground control points (GCPs) configuration on the 3D accuracy, and monocular depth estimation.

2 PAPERS • NO BENCHMARKS YET

Foot3D

A dataset of high resolution, textured scans of articulated left feet, useful for 3D shape representation learning.

2 PAPERS • NO BENCHMARKS YET

G-VUE (General-purpose Visual Understanding Evaluation)

General-purpose Visual Understanding Evaluation (G-VUE) is a comprehensive benchmark covering the full spectrum of visual cognitive abilities with four functional domains -- Perceive, Ground, Reason, and Act. The four domains are embodied in 11 carefully curated tasks, from 3D reconstruction to visual reasoning and manipulation.

2 PAPERS • NO BENCHMARKS YET

MBW - Zoo Dataset

Dataset page: https://github.com/mosamdabhi/MBW-Data

2 PAPERS • NO BENCHMARKS YET

Middlebury MVS

Middlebury MVS is the earliest MVS dataset for multi-view stereo network evaluation. It contains two indoor objects with low-resolution (640 × 480) images and calibrated cameras.

2 PAPERS • NO BENCHMARKS YET

Pano3D

Pano3D is a new benchmark for depth estimation from spherical panoramas. Its goal is to drive progress for this task in a consistent and holistic manner. The Pano3D 360 depth estimation benchmark provides a standard Matterport3D train and test split, as well as a secondary GibsonV2 partioning for testing and training as well. The latter is used for zero-shot cross dataset transfer performance assessment and decomposes it into 3 different splits, each one focusing on a specific generalization axis.

2 PAPERS • NO BENCHMARKS YET

Parcel3D

Synthetic dataset of over 13,000 images of damaged and intact parcels with full 2D and 3D annotations in the COCO format. For details see our paper and for visual samples our project page.

2 PAPERS • NO BENCHMARKS YET

AMT Objects

AMT Objects is a large dataset of object centric videos suitable for training and benchmarking models for generating 3D models of objects from a small number of photos of the objects. The dataset consists of multiple views of a large collection of object instances.

1 PAPER • NO BENCHMARKS YET

CSD
CSD (Collaborative SLAM Dataset)

Comprises 4 different subsets - Flat, House, Priory and Lab - each containing a number of different sequences that can be successfully relocalised against each other.

1 PAPER • NO BENCHMARKS YET

DRACO20K
DRACO20K

DRACO20K dataset is used for evaluating object canonicalization on methods that estimate a canonical frame from a monocular input image.

1 PAPER • NO BENCHMARKS YET

Danish Airs and Grounds
Danish Airs and Grounds

Danish Airs and Grounds (DAG) is a large collection of street-level and aerial images targeting such cases. Its main challenge lies in the extreme viewing-angle difference between query and reference images with consequent changes in illumination and perspective. The dataset is larger and more diverse than current publicly available data, including more than 50 km of road in urban, suburban and rural areas. All images are associated with accurate 6-DoF metadata that allows the benchmarking of visual localization methods.

1 PAPER • NO BENCHMARKS YET

Drunkard's Dataset

Estimating camera motion in deformable scenes poses a complex and open research challenge. Most existing non-rigid structure from motion techniques assume to observe also static scene parts besides deforming scene parts in order to establish an anchoring reference. However, this assumption does not hold true in certain relevant application cases such as endoscopies. To tackle this issue with a common benchmark, we introduce the Drunkard’s Dataset, a challenging collection of synthetic data targeting visual navigation and reconstruction in deformable environments. This dataset is the first large set of exploratory camera trajectories with ground truth inside 3D scenes where every surface exhibits non-rigid deformations over time. Simulations in realistic 3D buildings lets us obtain a vast amount of data and ground truth labels, including camera poses, RGB images and depth, optical flow and normal maps at high resolution and quality.

1 PAPER • 1 BENCHMARK

Indoor and outdoor DFD dataset

The dfd_indoor dataset contains 110 images for training and 29 images for testing. The dfd_outdoor dataset contains 34 images for tests; no ground truth was given for this dataset, as the depth sensor only works on indoor scenes.

1 PAPER • NO BENCHMARKS YET

Replay

Replay is a collection of multi-view, multi-modal videos of humans interacting socially. Each scene is filmed in high production quality, from different viewpoints with several static cameras, as well as wearable action cameras, and recorded with a large array of microphones at different positions in the room. The full Replay dataset consists of 68 scenes of social interactions between people, such as playing boarding games, exercising, or unwrapping presents. Each scene is about 5 minutes long and filmed with 12 cameras, static and dynamic. Audio is captured separately by 12 binaural microphones and additional near-range microphones for each actor and for each egocentric video. All sensors are temporally synchronized, undistorted, geometrically calibrated, and color calibrated.

1 PAPER • NO BENCHMARKS YET

Robust e-NeRF Synthetic Event Dataset
Robust e-NeRF Synthetic Event Dataset

This synthetic event dataset is used in Robust e-NeRF to study the collective effect of camera speed profile, contrast threshold variation and refractory period on the quality of NeRF reconstruction from a moving event camera. It is simulated using an improved version of ESIM with three different camera configurations of increasing difficulty levels (i.e. easy, medium and hard) on seven Realistic Synthetic $360^{\circ}$ scenes (adopted in the synthetic experiments of NeRF), resulting in a total of 21 sequence recordings. Please refer to the Robust e-NeRF paper for more details.

1 PAPER • NO BENCHMARKS YET

Contact us on: hello@paperswithcode.com . Papers With Code is a free resource with all data licensed under CC-BY-SA.
Terms Data policy Cookies policy