Ce Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Ce Liu has a total of 24 co-authors. Here are the top co-authors by number of shared publications:
- Bin Xiao
School of Artificial Intelligence, Chongqing University of Posts and Telecommunications, Chongqing 400065, China
1 shared publication - Bingsong Yu
School of Earth Sciences and Resources, China University of Geosciences, Beijing 100083, China
1 shared publication Chunyuan Li
1 shared publication- Cong Tan
PetroChina Research Institute of Petroleum Exploration & Development 1 , Beijing, China
1 shared publication Forrester Cole
Google research
1 shared publication- He Bi
Research Institute of Petroleum Exploration and Development 1 , Beijing, China
1 shared publication
Latest Publications
- 19 September 2023
- Published by Society of Exploration Geophysicists in Interpretation
- Vol. 12 (1) , T29-T46
Color origin and its sedimentary and paleoenvironmental significance of the Permian-Triassic strata in the Ordos Basin, China
The Permian-Triassic strata in the Ordos Basin deposit a colorful set of strata. Although there have been many studies on oil and gas resources here, little attention has been paid to the genesis mechanism of its color. To analyze the color origin of the Permian-Triassic claystones and to clarify the dialectical relationship between rock color and sedimentary environment, petrological and geochemical methods such as polarization microscope, scan electron microscope, X-ray diffractometer, and inductively coupled plasma-mass spectrometer are used. The results indicate that the minerals with different colors, composition, and content serve as colorants in sedimentary rocks, with the predominant dyeing component in dark claystones being the organic matter, whereas red beds primarily acquire their color from hematite and other ferric minerals. In addition, different rock colors correspond to distinct chemical composition variations. The black-gray rocks of the Yanchang and Shihezi Formations have high total organic carbon content, Fe2+/Fe3+, V/Cr, and low CaO/(MgO × Al2O3) and Sr/Cu, whereas the red-brown rocks of the Liujiagou and Heshanggou Formations exhibit the opposite. Therefore, different colors of the rocks stem from their different mineral and chemical compositions, which in turn influenced by the changes in the paleoclimate, paleoenvironment, and tectonic movements. Comprehensive analysis of color, mineral, and chemical composition reveals the evolution process of paleoclimate and paleoenvironment in the Ordos Basin from the late Permian to Triassic, beginning with a warm humid climate characterized by a weak oxidation environment in the Shihezi and Shiqianfeng Formations, transitioning to a hot arid climate in the Liujiagou and Heshanggou Formations, and returning to a warm humid climate with a weak oxidation environment in the Zhifang and Yanchang Formations. This evolution history aligns with the global tectonic and climate evolution. Overall, systematic analysis of sedimentary rock color can provide an important basis for the study of the paleoclimate and paleoenvironment.
- 19 September 2023
- Published by Society of Exploration Geophysicists in Interpretation
- Vol. 12 (1) , T29-T46
Color origin and its sedimentary and paleoenvironmental significance of the Permian-Triassic strata in the Ordos Basin, China
The Permian-Triassic strata in the Ordos Basin deposit a colorful set of strata. Although there have been many studies on oil and gas resources here, little attention has been paid to the genesis mechanism of its color. To analyze the color origin of the Permian-Triassic claystones and to clarify the dialectical relationship between rock color and sedimentary environment, petrological and geochemical methods such as polarization microscope, scan electron microscope, X-ray diffractometer, and inductively coupled plasma-mass spectrometer are used. The results indicate that the minerals with different colors, composition, and content serve as colorants in sedimentary rocks, with the predominant dyeing component in dark claystones being the organic matter, whereas red beds primarily acquire their color from hematite and other ferric minerals. In addition, different rock colors correspond to distinct chemical composition variations. The black-gray rocks of the Yanchang and Shihezi Formations have high total organic carbon content, Fe2+/Fe3+, V/Cr, and low CaO/(MgO × Al2O3) and Sr/Cu, whereas the red-brown rocks of the Liujiagou and Heshanggou Formations exhibit the opposite. Therefore, different colors of the rocks stem from their different mineral and chemical compositions, which in turn influenced by the changes in the paleoclimate, paleoenvironment, and tectonic movements. Comprehensive analysis of color, mineral, and chemical composition reveals the evolution process of paleoclimate and paleoenvironment in the Ordos Basin from the late Permian to Triassic, beginning with a warm humid climate characterized by a weak oxidation environment in the Shihezi and Shiqianfeng Formations, transitioning to a hot arid climate in the Liujiagou and Heshanggou Formations, and returning to a warm humid climate with a weak oxidation environment in the Zhifang and Yanchang Formations. This evolution history aligns with the global tectonic and climate evolution. Overall, systematic analysis of sedimentary rock color can provide an important basis for the study of the paleoclimate and paleoenvironment.
- Open Access
- 1 January 2023
- Published by The Econometric Society in Theoretical Economics
- Vol. 18 (4) , 1711-1757
Stability in repeated matching markets
This paper develops a framework for studying repeated matching markets. The model departs from the Gale–Shapley matching model by having a fixed set of long‐lived players (firms) match with a new generation of short‐lived players (workers) in every period. I define history‐dependent and self‐enforcing matching processes in this repeated matching environment and characterize the firms' payoffs. Firms fall into one of two categories: some firms must obtain the same payoff as they would in static stable matchings, and this holds at every patience level; meanwhile, repetition and history dependence can enlarge the set of sustainable payoffs for the other firms, provided that the firms are sufficiently patient. In large matching markets with correlated preferences, the first kind of firms corresponds to “elite” firms that make up at most a vanishingly small fraction of the market. The vast majority of firms fall into the second category.
- 1 January 2023
- Published by The Econometric Society in Theoretical Economics
- Vol. 18 (4) , 1711-1757
Open AccessStability in repeated matching markets
This paper develops a framework for studying repeated matching markets. The model departs from the Gale–Shapley matching model by having a fixed set of long‐lived players (firms) match with a new generation of short‐lived players (workers) in every period. I define history‐dependent and self‐enforcing matching processes in this repeated matching environment and characterize the firms' payoffs. Firms fall into one of two categories: some firms must obtain the same payoff as they would in static stable matchings, and this holds at every patience level; meanwhile, repetition and history dependence can enlarge the set of sustainable payoffs for the other firms, provided that the firms are sufficiently patient. In large matching markets with correlated preferences, the first kind of firms corresponds to “elite” firms that make up at most a vanishingly small fraction of the market. The vast majority of firms fall into the second category.
- Open Access
Unified Contrastive Learning in Image-Text-Label Space
Visual recognition is recently learned via either supervised learning on human-annotated image-label data or language-image contrastive learning with webly-crawled image-text pairs. While supervised learning may result in a more discriminative representation, language-image pretraining shows unprecedented zero-shot recognition capability, largely due to the different properties of data sources and learning objectives. In this work, we introduce a new formulation by combining the two data sources into a common image-text-label space. In this space, we propose a new learning paradigm, called Unified Contrastive Learning (UniCL) with a single learning objective to seamlessly prompt the synergy of two data types. Extensive experiments show that our UniCL is an effective way of learning semantically rich yet discriminative representations, universally for image recognition in zero-shot, linear-probe, fully finetuning and transfer learning scenarios. Particularly, it attains gains up to 9.2% and 14.5% in average on zero-shot recognition benchmarks over the language-image contrastive learning and supervised learning methods, respectively. In linear probe setting, it also boosts the performance over the two methods by 7.3% and 3.4%, respectively. Our study also indicates that UniCL stand-alone is a good learner on pure image-label data, rivaling the supervised learning methods across three image classification datasets and two types of vision backbones, ResNet and Swin Transformer. Code is available at https://github.com/microsoft/UniCL.
Open AccessUnified Contrastive Learning in Image-Text-Label Space
Visual recognition is recently learned via either supervised learning on human-annotated image-label data or language-image contrastive learning with webly-crawled image-text pairs. While supervised learning may result in a more discriminative representation, language-image pretraining shows unprecedented zero-shot recognition capability, largely due to the different properties of data sources and learning objectives. In this work, we introduce a new formulation by combining the two data sources into a common image-text-label space. In this space, we propose a new learning paradigm, called Unified Contrastive Learning (UniCL) with a single learning objective to seamlessly prompt the synergy of two data types. Extensive experiments show that our UniCL is an effective way of learning semantically rich yet discriminative representations, universally for image recognition in zero-shot, linear-probe, fully finetuning and transfer learning scenarios. Particularly, it attains gains up to 9.2% and 14.5% in average on zero-shot recognition benchmarks over the language-image contrastive learning and supervised learning methods, respectively. In linear probe setting, it also boosts the performance over the two methods by 7.3% and 3.4%, respectively. Our study also indicates that UniCL stand-alone is a good learner on pure image-label data, rivaling the supervised learning methods across three image classification datasets and two types of vision backbones, ResNet and Swin Transformer. Code is available at https://github.com/microsoft/UniCL.
- Open Access
Learning the Depths of Moving People by Watching Frozen People
We present a method for predicting dense depth in scenarios where both a monocular camera and people in the scene are freely moving. Existing methods for recovering depth for dynamic, non-rigid objects from monocular video impose strong assumptions on the objects' motion and may only recover sparse depth. In this paper, we take a data-driven approach and learn human depth priors from a new source of data: thousands of Internet videos of people imitating mannequins, i.e., freezing in diverse, natural poses, while a hand-held camera tours the scene. Because people are stationary, training data can be generated using multi-view stereo reconstruction. At inference time, our method uses motion parallax cues from the static areas of the scenes to guide the depth prediction. We demonstrate our method on real-world sequences of complex human actions captured by a moving hand-held camera, show improvement over state-of-the-art monocular depth prediction methods, and show various 3D effects produced using our predicted depth.
Open AccessLearning the Depths of Moving People by Watching Frozen People
We present a method for predicting dense depth in scenarios where both a monocular camera and people in the scene are freely moving. Existing methods for recovering depth for dynamic, non-rigid objects from monocular video impose strong assumptions on the objects' motion and may only recover sparse depth. In this paper, we take a data-driven approach and learn human depth priors from a new source of data: thousands of Internet videos of people imitating mannequins, i.e., freezing in diverse, natural poses, while a hand-held camera tours the scene. Because people are stationary, training data can be generated using multi-view stereo reconstruction. At inference time, our method uses motion parallax cues from the static areas of the scenes to guide the depth prediction. We demonstrate our method on real-world sequences of complex human actions captured by a moving hand-held camera, show improvement over state-of-the-art monocular depth prediction methods, and show various 3D effects produced using our predicted depth.
- 1 June 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 21607508,p. 429-436
Duplicate Discovery on 2 Billion Internet Images
Duplicate image discovery, or discovering duplicate image clusters, is a challenging problem for billions of Internet images due to the lack of good distance metric which both covers the large variation within a duplicate image cluster and eliminates false alarms. After carefully investigating existing local and global features that have been widely used for large-scale image search and indexing, we propose a two-step approach that combines both local and global features: global descriptors are used to discover seed clusters with high precision, whereas local descriptors are used to grow the seeds to cover good recall. Using efficient hashing techniques for both features and the MapReduce framework, our system is able to discover about 553.8 million duplicate images from 2 billion Internet images within 13 hours on a 2, 000 core cluster.
- 1 June 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 21607508,p. 429-436
Duplicate Discovery on 2 Billion Internet Images
Duplicate image discovery, or discovering duplicate image clusters, is a challenging problem for billions of Internet images due to the lack of good distance metric which both covers the large variation within a duplicate image cluster and eliminates false alarms. After carefully investigating existing local and global features that have been widely used for large-scale image search and indexing, we propose a two-step approach that combines both local and global features: global descriptors are used to discover seed clusters with high precision, whereas local descriptors are used to grow the seeds to cover good recall. Using efficient hashing techniques for both features and the MapReduce framework, our system is able to discover about 553.8 million duplicate images from 2 billion Internet images within 13 hours on a 2, 000 core cluster.
- Open Access
- 1 November 2011
- Published by Association for Computing Machinery (ACM)
- p. 477-480
A visual analytics system for metropolitan transportation
With the increasing availability of metropolitan transportation data, such as those from vehicle GPSs (Global Positioning Systems) and road-side sensors, it becomes viable for authorities, operators, as well as individuals to analyze the data for a better understanding of the transportation system and possibly improved utilization and planning of the system. We report our experience in building the VAST (Visual Analytics for Smart Transportation) system. Our key observation is that metropolitan transportation data are inherently visual as they are spatio-temporal around road networks. Therefore, we visualize traffic data together with digital maps and support analytical queries through this interactive visual interface. As a case study, we demonstrate VAST on real-world taxi GPS and meter data sets from 15, 000 taxis running two months in a Chinese city of over 10 million population. We discuss the technical challenges in data cleaning, storage, visualization, and query processing, and offer our first-hand lessons learned from developing the system. © 2011 Authors
- 1 November 2011
- Published by Association for Computing Machinery (ACM)
- p. 477-480
Open AccessA visual analytics system for metropolitan transportation
With the increasing availability of metropolitan transportation data, such as those from vehicle GPSs (Global Positioning Systems) and road-side sensors, it becomes viable for authorities, operators, as well as individuals to analyze the data for a better understanding of the transportation system and possibly improved utilization and planning of the system. We report our experience in building the VAST (Visual Analytics for Smart Transportation) system. Our key observation is that metropolitan transportation data are inherently visual as they are spatio-temporal around road networks. Therefore, we visualize traffic data together with digital maps and support analytical queries through this interactive visual interface. As a case study, we demonstrate VAST on real-world taxi GPS and meter data sets from 15, 000 taxis running two months in a Chinese city of over 10 million population. We discuss the technical challenges in data cleaning, storage, visualization, and query processing, and offer our first-hand lessons learned from developing the system. © 2011 Authors