In our research we test data and models for the recognition of housing quality in the city of Amsterdam from ground-level and aerial imagery. For ground-level images we compare Google StreetView (GSV) to Flickr images. Our results show that GSV predicts the most accurate building quality scores, approximately 30% better than using only aerial images. However, we find that through careful filtering and by using the right pre-trained model, Flickr image features combined with aerial image features are able to halve the performance gap to GSV features from 30% to 15%. Our results indicate that there are viable alternatives to GSV for liveability factor prediction, which is encouraging as GSV images are more difficult to acquire and not always available.
{"title":"Cross-Modal Learning of Housing Quality in Amsterdam","authors":"A. Levering, Diego Marcos, Ilan Havinga, D. Tuia","doi":"10.1145/3486635.3491067","DOIUrl":"https://doi.org/10.1145/3486635.3491067","url":null,"abstract":"In our research we test data and models for the recognition of housing quality in the city of Amsterdam from ground-level and aerial imagery. For ground-level images we compare Google StreetView (GSV) to Flickr images. Our results show that GSV predicts the most accurate building quality scores, approximately 30% better than using only aerial images. However, we find that through careful filtering and by using the right pre-trained model, Flickr image features combined with aerial image features are able to halve the performance gap to GSV features from 30% to 15%. Our results indicate that there are viable alternatives to GSV for liveability factor prediction, which is encouraging as GSV images are more difficult to acquire and not always available.","PeriodicalId":448866,"journal":{"name":"Proceedings of the 4th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128519591","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Many historical map sheets are publicly available for studies that require long-term historical geographic data. The cartographic design of these maps includes a combination of map symbols and text labels. Automatically reading text labels from map images could greatly speed up the map interpretation and helps generate rich metadata describing the map content. Many text detection algorithms have been proposed to locate text regions in map images automatically, but most of the algorithms are trained on out-of-domain datasets (e.g., scenic images). Training data determines the quality of machine learning models, and manually annotating text regions in map images is labor-extensive and time-consuming. On the other hand, existing geographic data sources, such as Open-StreetMap (OSM), contain machine-readable map layers, which allow us to separate out the text layer and obtain text label annotations easily. However, the cartographic styles between OSM map tiles and historical maps are significantly different. This paper proposes a method to automatically generate an unlimited amount of annotated historical map images for training text detection models. We use a style transfer model to convert contemporary map images into historical style and place text labels upon them. We show that the state-of-the-art text detection models (e.g., PSENet) can benefit from the synthetic historical maps and achieve significant improvement for historical map text detection.
{"title":"Synthetic Map Generation to Provide Unlimited Training Data for Historical Map Text Detection","authors":"Zekun Li, Runyu Guan, Qianmu Yu, Yao-Yi Chiang, Craig A. Knoblock","doi":"10.1145/3486635.3491070","DOIUrl":"https://doi.org/10.1145/3486635.3491070","url":null,"abstract":"Many historical map sheets are publicly available for studies that require long-term historical geographic data. The cartographic design of these maps includes a combination of map symbols and text labels. Automatically reading text labels from map images could greatly speed up the map interpretation and helps generate rich metadata describing the map content. Many text detection algorithms have been proposed to locate text regions in map images automatically, but most of the algorithms are trained on out-of-domain datasets (e.g., scenic images). Training data determines the quality of machine learning models, and manually annotating text regions in map images is labor-extensive and time-consuming. On the other hand, existing geographic data sources, such as Open-StreetMap (OSM), contain machine-readable map layers, which allow us to separate out the text layer and obtain text label annotations easily. However, the cartographic styles between OSM map tiles and historical maps are significantly different. This paper proposes a method to automatically generate an unlimited amount of annotated historical map images for training text detection models. We use a style transfer model to convert contemporary map images into historical style and place text labels upon them. We show that the state-of-the-art text detection models (e.g., PSENet) can benefit from the synthetic historical maps and achieve significant improvement for historical map text detection.","PeriodicalId":448866,"journal":{"name":"Proceedings of the 4th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128062318","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Automating post-disaster damage assessment with remote sensing data is critical for faster surveys of structures impacted by natural disasters. One significant obstacle to training state-of-the-art deep neural networks to support this automation is that large quantities of labelled data are often required. However, obtaining those labels is particularly unrealistic to support post-disaster damage assessment in a timely manner. Few-shot learning methods could help to mitigate this by reducing the amount of labelled data required to successfully train a model while achieving satisfactory results. To this end, we explore a feature reweighting method to the YOLOv3 object detection architecture to achieve few-shot learning of damage assessment models on the xBD dataset. Our results show that the feature reweighting approach yield improved mAP over the baseline with significantly fewer labelled samples. In addition, we use t-SNE to analyze the class-specific reweighting vectors generated by the reweighting module in order to evaluate their inter-class and intra-class similarity. We find that the vectors form clusters based on class, and that these clusters overlap with visually similar classes. Those results show the potential to employ this few-shot learning strategy for rapid damage assessment with post-event remote sensing images.
{"title":"Few-shot Learning for Post-disaster Structure Damage Assessment","authors":"Jordan Bowman, Lexie Yang","doi":"10.1145/3486635.3491071","DOIUrl":"https://doi.org/10.1145/3486635.3491071","url":null,"abstract":"Automating post-disaster damage assessment with remote sensing data is critical for faster surveys of structures impacted by natural disasters. One significant obstacle to training state-of-the-art deep neural networks to support this automation is that large quantities of labelled data are often required. However, obtaining those labels is particularly unrealistic to support post-disaster damage assessment in a timely manner. Few-shot learning methods could help to mitigate this by reducing the amount of labelled data required to successfully train a model while achieving satisfactory results. To this end, we explore a feature reweighting method to the YOLOv3 object detection architecture to achieve few-shot learning of damage assessment models on the xBD dataset. Our results show that the feature reweighting approach yield improved mAP over the baseline with significantly fewer labelled samples. In addition, we use t-SNE to analyze the class-specific reweighting vectors generated by the reweighting module in order to evaluate their inter-class and intra-class similarity. We find that the vectors form clusters based on class, and that these clusters overlap with visually similar classes. Those results show the potential to employ this few-shot learning strategy for rapid damage assessment with post-event remote sensing images.","PeriodicalId":448866,"journal":{"name":"Proceedings of the 4th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117177661","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Trajectory data is among the most sensitive data and the society increasingly raises privacy concerns. In this demo paper, we present a privacy-preserving Vehicle Trajectory Simulation and Visualization (VTSV) web platform (demo video: https://youtu.be/NY5L4bu2kTU), which automatically generates navigation routes between given pairs of origins and destinations and employs a deep reinforcement learning model to simulate vehicle trajectories with customized driving behaviors such as normal driving, overspeed, aggressive acceleration, and aggressive turning. The simulated vehicle trajectory data contain high-sample-rate of attributes including GPS location, speed, acceleration, and steering angle, and such data are visualized in VTSV using streetscape.gl, an autonomous driving data visualization framework. Location privacy protection methods such as origin-destination geomasking and trajectory k-anonymity are integrated into the platform to support privacy-preserving trajectory data generation and publication. We design two application scenarios to demonstrate how VTSV performs location privacy protection and customize driving behavior, respectively. The demonstration shows that VTSV is able to mitigate data privacy, sparsity, and imbalance sampling issues, which offers new insights into driving trajectory simulation and GeoAI-powered privacy-preserving data publication.
{"title":"VTSV","authors":"Jinmeng Rao, Song Gao, Xiaojin Zhu","doi":"10.1145/3486635.3491073","DOIUrl":"https://doi.org/10.1145/3486635.3491073","url":null,"abstract":"Trajectory data is among the most sensitive data and the society increasingly raises privacy concerns. In this demo paper, we present a privacy-preserving Vehicle Trajectory Simulation and Visualization (VTSV) web platform (demo video: https://youtu.be/NY5L4bu2kTU), which automatically generates navigation routes between given pairs of origins and destinations and employs a deep reinforcement learning model to simulate vehicle trajectories with customized driving behaviors such as normal driving, overspeed, aggressive acceleration, and aggressive turning. The simulated vehicle trajectory data contain high-sample-rate of attributes including GPS location, speed, acceleration, and steering angle, and such data are visualized in VTSV using streetscape.gl, an autonomous driving data visualization framework. Location privacy protection methods such as origin-destination geomasking and trajectory k-anonymity are integrated into the platform to support privacy-preserving trajectory data generation and publication. We design two application scenarios to demonstrate how VTSV performs location privacy protection and customize driving behavior, respectively. The demonstration shows that VTSV is able to mitigate data privacy, sparsity, and imbalance sampling issues, which offers new insights into driving trajectory simulation and GeoAI-powered privacy-preserving data publication.","PeriodicalId":448866,"journal":{"name":"Proceedings of the 4th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125577245","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Md. Mostafijur Rahman, Arpan Man Sainju, Dan Yan, Zhe Jiang
Road safety barriers (e.g., concrete barriers, metal crash barriers, rumble strips) play an important role in preventing or mitigating vehicle crashes. Accurate maps of road safety barriers are critical components of safety infrastructure management systems at federal or state transportation agencies. In current practice, mapping road safety barriers is largely done manually (e.g., driving on the road or visual interpretation of street view imagery), which is slow, tedious, and expensive. We propose a deep learning approach to automatically map road safety barriers from street view imagery. Our approach considers road barriers as long objects spanning across consecutive street view images in a sequence and use a hybrid object-detection and recurrent-network model. Preliminary results on real-world street view imagery show that the proposed model outperforms several baseline methods.
{"title":"Mapping Road Safety Barriers Across Street View Image Sequences: A Hybrid Object Detection and Recurrent Model","authors":"Md. Mostafijur Rahman, Arpan Man Sainju, Dan Yan, Zhe Jiang","doi":"10.1145/3486635.3491074","DOIUrl":"https://doi.org/10.1145/3486635.3491074","url":null,"abstract":"Road safety barriers (e.g., concrete barriers, metal crash barriers, rumble strips) play an important role in preventing or mitigating vehicle crashes. Accurate maps of road safety barriers are critical components of safety infrastructure management systems at federal or state transportation agencies. In current practice, mapping road safety barriers is largely done manually (e.g., driving on the road or visual interpretation of street view imagery), which is slow, tedious, and expensive. We propose a deep learning approach to automatically map road safety barriers from street view imagery. Our approach considers road barriers as long objects spanning across consecutive street view images in a sequence and use a hybrid object-detection and recurrent-network model. Preliminary results on real-world street view imagery show that the proposed model outperforms several baseline methods.","PeriodicalId":448866,"journal":{"name":"Proceedings of the 4th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134403327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rutuja Gurav, Debraj De, Gautam S. Thakur, Junchuan Fan
With the proliferation of smartphone cameras and social networks, we have rich, multi-modal data about points of interest (POIs) - like cultural landmarks, institutions, businesses, etc. - within a given areas of interest (AOI) (e.g., a county, city or a neighborhood) available to us. Data conflation across multiple modalities of data sources is one of the key challenges in maintaining a geographical information system (GIS) which accumulate data about POIs. Given POI data from nine different sources, and ground-level geo-tagged and scene-captioned images from two different image hosting platforms, in this work we explore the application of graph neural networks (GNNs) to perform data conflation, while leveraging a natural graph structure evident in geospatial data. The preliminary results demonstrate the capacity of a GNN operation to learn distributions of entity (POIs and images) features, coupled with topological structure of entity's local neighborhood in a semantic nearest neighbor graph, in order to predict links between a pair of entities.
{"title":"Conflation of Geospatial POI Data and Ground-level Imagery via Link Prediction on Joint Semantic Graph","authors":"Rutuja Gurav, Debraj De, Gautam S. Thakur, Junchuan Fan","doi":"10.1145/3486635.3491068","DOIUrl":"https://doi.org/10.1145/3486635.3491068","url":null,"abstract":"With the proliferation of smartphone cameras and social networks, we have rich, multi-modal data about points of interest (POIs) - like cultural landmarks, institutions, businesses, etc. - within a given areas of interest (AOI) (e.g., a county, city or a neighborhood) available to us. Data conflation across multiple modalities of data sources is one of the key challenges in maintaining a geographical information system (GIS) which accumulate data about POIs. Given POI data from nine different sources, and ground-level geo-tagged and scene-captioned images from two different image hosting platforms, in this work we explore the application of graph neural networks (GNNs) to perform data conflation, while leveraging a natural graph structure evident in geospatial data. The preliminary results demonstrate the capacity of a GNN operation to learn distributions of entity (POIs and images) features, coupled with topological structure of entity's local neighborhood in a semantic nearest neighbor graph, in order to predict links between a pair of entities.","PeriodicalId":448866,"journal":{"name":"Proceedings of the 4th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129415555","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Semantic segmentation using deep neural networks is an important component of aerial image understanding. However, models trained using data from one domain may not generalize well to another domain due to a domain shift between data distributions in the two domains. Such a domain gap is common in aerial images due to large visual appearance changes, and so substantial accuracy loss may occur when using a trained model for inference on new data. In this paper, we propose a novel unsupervised domain adaptation framework to address domain shift in the context of semantic segmentation of aerial images. To this end, we address the problem of domain shift by learning class-aware distribution differences between the source and target domains. Further, we employ entropy minimization on the target domain to produce high-confidence predictions. We demonstrate the effectiveness of the proposed approach using a challenge segmentation dataset by ISPRS, and show improvement over state-of-the-art methods.
{"title":"Semantic Segmentation in Aerial Images Using Class-Aware Unsupervised Domain Adaptation","authors":"Ying Chen, Xu Ouyang, Kaiyue Zhu, G. Agam","doi":"10.1145/3486635.3491069","DOIUrl":"https://doi.org/10.1145/3486635.3491069","url":null,"abstract":"Semantic segmentation using deep neural networks is an important component of aerial image understanding. However, models trained using data from one domain may not generalize well to another domain due to a domain shift between data distributions in the two domains. Such a domain gap is common in aerial images due to large visual appearance changes, and so substantial accuracy loss may occur when using a trained model for inference on new data. In this paper, we propose a novel unsupervised domain adaptation framework to address domain shift in the context of semantic segmentation of aerial images. To this end, we address the problem of domain shift by learning class-aware distribution differences between the source and target domains. Further, we employ entropy minimization on the target domain to produce high-confidence predictions. We demonstrate the effectiveness of the proposed approach using a challenge segmentation dataset by ISPRS, and show improvement over state-of-the-art methods.","PeriodicalId":448866,"journal":{"name":"Proceedings of the 4th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131841638","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Location classification is used for associating type to locations, to enrich maps and support a plethora of geospatial applications that rely on location types. Classification can be performed by humans, but using machine learning is more efficient and faster to react to changes than human-based classification. Machine learning can be used in lieu of human classification or for supporting it. In this paper we study the use of machine learning for Geosocial Location Classification, where the type of a site, e.g., a building, is discovered based on social-media posts, e.g., tweets. Our goal is to correctly associate a set of tweets posted in a small radius around a given location with the corresponding location type, e.g., school, church, restaurant or museum. We explore two approaches to the problem: (a) a pipeline approach, where each post is first classified, and then the location associated with the set of posts is inferred from the individual post labels; and (b) a joint approach where the individual posts are simultaneously processed to yield the desired location type. We tested the two approaches over a data set of geotagged tweets. Our results demonstrate the superiority of the joint approach. Moreover, we show that due to the unique structure of the problem, where weakly-related messages are jointly processed to yield a single final label, linear classifiers outperform deep neural network alternatives.
{"title":"Location Classification Based on Tweets","authors":"Elad Kravi, Y. Kanza, B. Kimelfeld, Roi Reichart","doi":"10.1145/3486635.3491075","DOIUrl":"https://doi.org/10.1145/3486635.3491075","url":null,"abstract":"Location classification is used for associating type to locations, to enrich maps and support a plethora of geospatial applications that rely on location types. Classification can be performed by humans, but using machine learning is more efficient and faster to react to changes than human-based classification. Machine learning can be used in lieu of human classification or for supporting it. In this paper we study the use of machine learning for Geosocial Location Classification, where the type of a site, e.g., a building, is discovered based on social-media posts, e.g., tweets. Our goal is to correctly associate a set of tweets posted in a small radius around a given location with the corresponding location type, e.g., school, church, restaurant or museum. We explore two approaches to the problem: (a) a pipeline approach, where each post is first classified, and then the location associated with the set of posts is inferred from the individual post labels; and (b) a joint approach where the individual posts are simultaneously processed to yield the desired location type. We tested the two approaches over a data set of geotagged tweets. Our results demonstrate the superiority of the joint approach. Moreover, we show that due to the unique structure of the problem, where weakly-related messages are jointly processed to yield a single final label, linear classifiers outperform deep neural network alternatives.","PeriodicalId":448866,"journal":{"name":"Proceedings of the 4th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130874416","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Representation learning of spatial and geographic data is a rapidly developing field which allows for similarity detection between areas and high-quality inference using deep neural networks. Past approaches however concentrated on embedding raster imagery (maps, street or satellite photos), mobility data or road networks. In this paper we propose the first approach to learning vector representations of OpenStreetMap regions with respect to urban functions and land-use in a micro-region grid. We identify a subset of OSM tags related to major characteristics of land-use, building and urban region functions, types of water, green or other natural areas. Through manual verification of tagging quality, we selected 36 cities were for training region representations. Uber's H3 index was used to divide the cities into hexagons, and OSM tags were aggregated for each hexagon. We propose the hex2vec method based on the Skip-gram model with negative sampling. The resulting vector representations showcase semantic structures of the map characteristics, similar to ones found in vector-based language models. We also present insights from region similarity detection in six Polish cities and propose a region typology obtained through agglomerative clustering.
{"title":"hex2vec: Context-Aware Embedding H3 Hexagons with OpenStreetMap Tags","authors":"Szymon Wo'zniak, Piotr Szyma'nski","doi":"10.1145/3486635.3491076","DOIUrl":"https://doi.org/10.1145/3486635.3491076","url":null,"abstract":"Representation learning of spatial and geographic data is a rapidly developing field which allows for similarity detection between areas and high-quality inference using deep neural networks. Past approaches however concentrated on embedding raster imagery (maps, street or satellite photos), mobility data or road networks. In this paper we propose the first approach to learning vector representations of OpenStreetMap regions with respect to urban functions and land-use in a micro-region grid. We identify a subset of OSM tags related to major characteristics of land-use, building and urban region functions, types of water, green or other natural areas. Through manual verification of tagging quality, we selected 36 cities were for training region representations. Uber's H3 index was used to divide the cities into hexagons, and OSM tags were aggregated for each hexagon. We propose the hex2vec method based on the Skip-gram model with negative sampling. The resulting vector representations showcase semantic structures of the map characteristics, similar to ones found in vector-based language models. We also present insights from region similarity detection in six Polish cities and propose a region typology obtained through agglomerative clustering.","PeriodicalId":448866,"journal":{"name":"Proceedings of the 4th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128750671","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
C. V. K. Iyer, Feili Hou, Henry Wang, Yonghong Wang, Kay Oh, Swetava Ganguli, Vipul Pandey
We present a no-code Artificial Intelligence (AI) platform called Trinity with the main design goal of enabling both machine learning researchers and non-technical geospatial domain experts to experiment with domain-specific signals and datasets for solving a variety of complex problems on their own. This versatility to solve diverse problems is achieved by transforming complex Spatio-temporal datasets to make them consumable by standard deep learning models, in this case, Convolutional Neural Networks (CNNs), and giving the ability to formulate disparate problems in a standard way, eg. semantic segmentation. With an intuitive user interface, a feature store that hosts derivatives of complex feature engineering, a deep learning kernel, and a scalable data processing mechanism, Trinity provides a powerful platform for domain experts to share the stage with scientists and engineers in solving business-critical problems. It enables quick prototyping, rapid experimentation and reduces the time to production by standardizing model building and deployment. In this paper, we present our motivation behind Trinity and its design along with showcasing sample applications to motivate the idea of lowering the bar to using AI.
{"title":"Trinity: A No-Code AI platform for complex spatial datasets","authors":"C. V. K. Iyer, Feili Hou, Henry Wang, Yonghong Wang, Kay Oh, Swetava Ganguli, Vipul Pandey","doi":"10.1145/3486635.3491072","DOIUrl":"https://doi.org/10.1145/3486635.3491072","url":null,"abstract":"We present a no-code Artificial Intelligence (AI) platform called Trinity with the main design goal of enabling both machine learning researchers and non-technical geospatial domain experts to experiment with domain-specific signals and datasets for solving a variety of complex problems on their own. This versatility to solve diverse problems is achieved by transforming complex Spatio-temporal datasets to make them consumable by standard deep learning models, in this case, Convolutional Neural Networks (CNNs), and giving the ability to formulate disparate problems in a standard way, eg. semantic segmentation. With an intuitive user interface, a feature store that hosts derivatives of complex feature engineering, a deep learning kernel, and a scalable data processing mechanism, Trinity provides a powerful platform for domain experts to share the stage with scientists and engineers in solving business-critical problems. It enables quick prototyping, rapid experimentation and reduces the time to production by standardizing model building and deployment. In this paper, we present our motivation behind Trinity and its design along with showcasing sample applications to motivate the idea of lowering the bar to using AI.","PeriodicalId":448866,"journal":{"name":"Proceedings of the 4th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130987992","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：