Tian Xing
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Tian Xing has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Qianchuan Zhao
Tsinghua University
11 shared publications - Kailai Sun
Singapore-MIT Alliance for Research and Technology Centre (SMART) Singapore
9 shared publications Xinwei Wang
Department of Automation, Center for Intelligent and Networked Systems, Tsinghua University, Beijing, China
4 shared publications- Hu Yan
Department of Automation, Center for Intelligent and Networked Systems, Tsinghua University, Beijing, China
3 shared publications Han Yu
Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Changchun, China
2 shared publicationsHu Yan
2 shared publications
Latest Publications
- Open Access
- 31 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 18 (19391404) , 28709-28725
DLDC: A Dual Loop Data Cleaning Method for Fine-Tuning Remote Sensing Image Generative Models
Text-to-image (T2I) generation, offering flexible and intuitive synthetic data for downstream geoscience applications, has garnered increasing attention in recent years. Training a good T2I model often requires high-quality, large-scale image text datasets. However, obtaining these datasets in remote sensing (RS) is challenging because of high annotation costs and specific domain knowledge. This study proposes a dual loop data cleaning (DLDC) method, which leverages contrastive multimodal quality evaluations to generate high-quality RS image-text training data automatically. By constructing an external generation loop (EGL) based on a multimodal foundational model and an internal evalu ation loop (IEL) based on contrastive learning metrics, DLDC can automatically generate layout description and evaluate image text matching degree on satellite images. The proposed approach effectively filters out noisy samples and curates a refined dataset without human intervention. Experimental results show that our dual loop evaluation can accurately determine the optimal data cleaning ratio for different scenes, improving image generation quality. Compared with the pre-trained T2I models, our fine tuned models reduce FID values by over 35%, increase CLIP scores by more than 25%, and improve RemoteCLIP scores by over 10.5%. Furthermore, our DLDC method can achieve superior performance compared to other state-of-the-art (SOTA) RS T2I models (e.g., Crs-diff, GeoRSSD, DiffusionSAT). Our data-cleaning method can improve downstream segmentation tasks, resulting in 8.14% in mean IoU and 7.5% in mean accuracy compared to the same model trained on raw or uncleaned data. Experimental results demonstrate that our automatically generated image-text data is of a similar quality to human manually annotated data, opening new pathways for rapid, cost effective, and reliable RS data generation.
- 31 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 18 (19391404) , 28709-28725
Open AccessDLDC: A Dual Loop Data Cleaning Method for Fine-Tuning Remote Sensing Image Generative Models
Text-to-image (T2I) generation, offering flexible and intuitive synthetic data for downstream geoscience applications, has garnered increasing attention in recent years. Training a good T2I model often requires high-quality, large-scale image text datasets. However, obtaining these datasets in remote sensing (RS) is challenging because of high annotation costs and specific domain knowledge. This study proposes a dual loop data cleaning (DLDC) method, which leverages contrastive multimodal quality evaluations to generate high-quality RS image-text training data automatically. By constructing an external generation loop (EGL) based on a multimodal foundational model and an internal evalu ation loop (IEL) based on contrastive learning metrics, DLDC can automatically generate layout description and evaluate image text matching degree on satellite images. The proposed approach effectively filters out noisy samples and curates a refined dataset without human intervention. Experimental results show that our dual loop evaluation can accurately determine the optimal data cleaning ratio for different scenes, improving image generation quality. Compared with the pre-trained T2I models, our fine tuned models reduce FID values by over 35%, increase CLIP scores by more than 25%, and improve RemoteCLIP scores by over 10.5%. Furthermore, our DLDC method can achieve superior performance compared to other state-of-the-art (SOTA) RS T2I models (e.g., Crs-diff, GeoRSSD, DiffusionSAT). Our data-cleaning method can improve downstream segmentation tasks, resulting in 8.14% in mean IoU and 7.5% in mean accuracy compared to the same model trained on raw or uncleaned data. Experimental results demonstrate that our automatically generated image-text data is of a similar quality to human manually annotated data, opening new pathways for rapid, cost effective, and reliable RS data generation.
- 5 August 2025
- Published by Elsevier in Energy and Buildings
- Vol. 347
Topology-aware hypergraph reinforcement learning for indoor occupant-centric HVAC control
- 5 August 2025
- Published by Elsevier in Energy and Buildings
- Vol. 347
Topology-aware hypergraph reinforcement learning for indoor occupant-centric HVAC control
- Open Access
- 20 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
DLDC: A Dual Loop Data Cleaning Method for Fine-tuning Remote Sensing Image Generative Models
Text-to-image (T2I) generation has gained increasing attention as it provides flexible and intuitive synthetic data for downstream geoscience applications. However, obtaining large-scale, high-quality image-text datasets in remote sensing (RS) is challenging because of high annotation costs and specialized domain knowledge. In this paper, we propose a dual loop data cleaning (DLDC) method, which leverages contrastive multimodal quality evaluations to generate high-quality RS image-text training data automatically. By constructing an external generation loop (EGL) based on a multimodal basic model and an internal evaluation loop (IEL) based on contrastive learning metrics, DLDC can automatically perform layout description and image-text matching evaluation on remote sensing images. Our approach effectively filters out noisy samples and curates a refined dataset without manual intervention. Experimental results show that our dual loop evaluation can accurately determine the optimal data cleaning ratio for different scenes, improving image generation quality. Compared with state-of-the-art (SOTA) pre-trained models, our fine-tuned models reduce FID values by over 35%, increase CLIP scores by more than 25%, and improve RemoteCLIP scores by over 10.5%. Experimental results demonstrate that our automatically generated image-text data is of a similar quality to manually annotated data, opening new pathways for rapid, cost-effective, and reliable RS data generation.
- 20 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Open AccessDLDC: A Dual Loop Data Cleaning Method for Fine-tuning Remote Sensing Image Generative Models
Text-to-image (T2I) generation has gained increasing attention as it provides flexible and intuitive synthetic data for downstream geoscience applications. However, obtaining large-scale, high-quality image-text datasets in remote sensing (RS) is challenging because of high annotation costs and specialized domain knowledge. In this paper, we propose a dual loop data cleaning (DLDC) method, which leverages contrastive multimodal quality evaluations to generate high-quality RS image-text training data automatically. By constructing an external generation loop (EGL) based on a multimodal basic model and an internal evaluation loop (IEL) based on contrastive learning metrics, DLDC can automatically perform layout description and image-text matching evaluation on remote sensing images. Our approach effectively filters out noisy samples and curates a refined dataset without manual intervention. Experimental results show that our dual loop evaluation can accurately determine the optimal data cleaning ratio for different scenes, improving image generation quality. Compared with state-of-the-art (SOTA) pre-trained models, our fine-tuned models reduce FID values by over 35%, increase CLIP scores by more than 25%, and improve RemoteCLIP scores by over 10.5%. Experimental results demonstrate that our automatically generated image-text data is of a similar quality to manually annotated data, opening new pathways for rapid, cost-effective, and reliable RS data generation.
- 6 July 2024
- Published by Elsevier in Energy and Buildings
- Vol. 319
High-accuracy occupancy counting at crowded entrances for smart buildings
- 6 July 2024
- Published by Elsevier in Energy and Buildings
- Vol. 319
High-accuracy occupancy counting at crowded entrances for smart buildings
- 7 September 2023
- Published by Elsevier in Building and Environment
- Vol. 244
Building occupancy number prediction: A Transformer approach
- 7 September 2023
- Published by Elsevier in Building and Environment
- Vol. 244
Building occupancy number prediction: A Transformer approach
- Open Access
Multi-Sensor-Based Occupancy Prediction in a Multi-Zone Office Building with Transformer
Buildings are responsible for approximately 40% of the world’s energy consumption and 36% of the total carbon dioxide emissions. Building occupancy is essential, enabling occupant-centric control for zero emissions and decarbonization. Although existing machine learning and deep learning methods for building occupancy prediction have made notable progress, their analyses remain limited when applied to complex real-world scenarios. Moreover, there is a high expectation for Transformer algorithms to predict building occupancy accurately. Therefore, this paper presents an occupancy prediction Transformer network (OPTnet). We fused and fed multi-sensor data (building occupancy, indoor environmental conditions, HVAC operations) into a Transformer model to forecast the future occupancy presence in multiple zones. We performed experimental analyses and compared it to different occupancy prediction methods (e.g., decision tree, long short-term memory networks, multi-layer perceptron) and diverse time horizons (1, 2, 3, 5, 10, 20, 30 min). Performance metrics (e.g., accuracy and mean squared error) were employed to evaluate the effectiveness of the prediction algorithms. Our OPTnet method achieved superior performance on our experimental two-week data compared to existing methods. The improved performance indicates its potential to enhance HVAC control systems and energy optimization strategies.
Open AccessMulti-Sensor-Based Occupancy Prediction in a Multi-Zone Office Building with Transformer
Buildings are responsible for approximately 40% of the world’s energy consumption and 36% of the total carbon dioxide emissions. Building occupancy is essential, enabling occupant-centric control for zero emissions and decarbonization. Although existing machine learning and deep learning methods for building occupancy prediction have made notable progress, their analyses remain limited when applied to complex real-world scenarios. Moreover, there is a high expectation for Transformer algorithms to predict building occupancy accurately. Therefore, this paper presents an occupancy prediction Transformer network (OPTnet). We fused and fed multi-sensor data (building occupancy, indoor environmental conditions, HVAC operations) into a Transformer model to forecast the future occupancy presence in multiple zones. We performed experimental analyses and compared it to different occupancy prediction methods (e.g., decision tree, long short-term memory networks, multi-layer perceptron) and diverse time horizons (1, 2, 3, 5, 10, 20, 30 min). Performance metrics (e.g., accuracy and mean squared error) were employed to evaluate the effectiveness of the prediction algorithms. Our OPTnet method achieved superior performance on our experimental two-week data compared to existing methods. The improved performance indicates its potential to enhance HVAC control systems and energy optimization strategies.
- Open Access
- 26 June 2023
- Published by MDPI AG in Preprints.org
Multi-sensor Based Occupancy Prediction in a Multi-zone Office Building with Transformer
Buildings are responsible for approximately 40% of the world’s energy consumption and 36% of the total carbon dioxide emissions. Building occupancy is essential, enabling Occupant-Centric Control for zero emissions and decarbonization. Although existing machine learning and deep learning methods for building occupancy prediction have achieved remarked progress, their analyses remain limited when applied to complex real-world scenarios. Besides, there is a high expectation for Transformer algorithms to predict building occupancy accurately. Therefore, this paper presents an Occupancy Prediction Transformer network (OPTnet). We fuse and feed multi-sensor data (building occupancy, indoor environmental conditions, HVAC operations) into a Transformer model to forecast the future occupancy presence in multiple zones. We perform experimental analysis and compare it to different occupancy prediction methods (e.g., Decision Tree, Long Short-Term Memory networks, Multi-layer perceptron) and diverse time horizons (1,2,3,5,10,20,30mins). The performance metrics (e.g., Accuracy and Mean Squared Error) are employed to evaluate the effectiveness of prediction algorithms. Our OPTnet method achieves superior performance on our experimental two-week data compared to existing methods. The improved performance signifies its potential to enhance HVAC control systems and energy optimization strategies. We will make the code publicly available at https://github.com/kailaisun/occupancy-prediction-binary to promote transparency and reproducibility.
- 26 June 2023
- Published by MDPI AG in Preprints.org
Open AccessMulti-sensor Based Occupancy Prediction in a Multi-zone Office Building with Transformer
Buildings are responsible for approximately 40% of the world’s energy consumption and 36% of the total carbon dioxide emissions. Building occupancy is essential, enabling Occupant-Centric Control for zero emissions and decarbonization. Although existing machine learning and deep learning methods for building occupancy prediction have achieved remarked progress, their analyses remain limited when applied to complex real-world scenarios. Besides, there is a high expectation for Transformer algorithms to predict building occupancy accurately. Therefore, this paper presents an Occupancy Prediction Transformer network (OPTnet). We fuse and feed multi-sensor data (building occupancy, indoor environmental conditions, HVAC operations) into a Transformer model to forecast the future occupancy presence in multiple zones. We perform experimental analysis and compare it to different occupancy prediction methods (e.g., Decision Tree, Long Short-Term Memory networks, Multi-layer perceptron) and diverse time horizons (1,2,3,5,10,20,30mins). The performance metrics (e.g., Accuracy and Mean Squared Error) are employed to evaluate the effectiveness of prediction algorithms. Our OPTnet method achieves superior performance on our experimental two-week data compared to existing methods. The improved performance signifies its potential to enhance HVAC control systems and energy optimization strategies. We will make the code publicly available at https://github.com/kailaisun/occupancy-prediction-binary to promote transparency and reproducibility.
- 11 May 2023
- Published by Elsevier in Building and Environment
- Vol. 239
MITP-Net: A deep-learning framework for short-term indoor temperature predictions in multi-zone buildings
- 11 May 2023
- Published by Elsevier in Building and Environment
- Vol. 239
MITP-Net: A deep-learning framework for short-term indoor temperature predictions in multi-zone buildings
- 28 September 2022
- Published by Elsevier in Building and Environment
- Vol. 225
A fusion framework for vision-based indoor occupancy estimation
- 28 September 2022
- Published by Elsevier in Building and Environment
- Vol. 225
A fusion framework for vision-based indoor occupancy estimation