• Corpus ID: 245218930

IS-COUNT: Large-scale Object Counting from Satellite Images with Covariate-based Importance Sampling

  title={IS-COUNT: Large-scale Object Counting from Satellite Images with Covariate-based Importance Sampling},
  author={Chenlin Meng and Enci Liu and Willie Neiswanger and Jiaming Song and M. Burke and D. Lobell and Stefano Ermon},
Object detection in high-resolution satellite imagery is emerging as a scalable alternative to on-the-ground survey data collection in many environmental and socioeconomic monitoring applications. However, performing object detection over large geographies can still be prohibitively expensive due to the high cost of purchasing imagery and compute. Inspired by traditional survey data collection strategies, we propose an approach to estimate object count statistics over large geographies through… 


Counting Dense Objects in Remote Sensing Images
This paper constructs a large-scale object counting dataset based on remote sensing images, which contains four kinds of objects and designs a novel neural network which can generate density map of an input image.
Efficient Poverty Mapping from High Resolution Remote Sensing Images
A reinforcement learning approach is proposed in which free lowresolution imagery is used to dynamically identify where to acquire costly high- resolution images, prior to performing a deep learning task on the high-resolution images.
xView: Objects in Context in Overhead Imagery
This satellite imagery dataset enables research progress pertaining to four key computer vision frontiers and utilizes a novel process for geospatial category detection and bounding box annotation with three stages of quality control.
Efficient Object Detection in Large Images Using Deep Reinforcement Learning
A reinforcement learning agent is proposed that adaptively selects the spatial resolution of each image that is provided to the detector to reduce the dependency on high spatial resolution images for building a robust detector and increases run-time efficiency.
Combining satellite imagery and machine learning to predict poverty
This work shows how a convolutional neural network can be trained to identify image features that can explain up to 75% of the variation in local-level economic outcomes, and could transform efforts to track and target poverty in developing countries.
Using publicly available satellite imagery and deep learning to understand economic well-being in Africa
Deep learning models are shown to be able to explain 70% of the variation in ground-measured village wealth in held-out countries, outperforming previous benchmarks from high-resolution imagery with errors comparable to that of existing ground data.
An automatic method for counting olive trees in very high spatial remote sensing images
An automatic method for counting olive trees in very high spatial remote sensing images using a Gaussian process classifier and the number of blobs in the image representing possible olive trees is counted using an automatic procedure.
Learning to Interpret Satellite Images Using Wikipedia
This work constructs a novel large-scale, multi-modal dataset by pairing geo-referenced Wikipedia articles with satellite imagery of their corresponding locations and demonstrates that this weak form of supervision can drastically reduce the quantity of human annotated labels and time required for downstream tasks.
Using Satellite Imagery to Understand and Promote Sustainable Development
The paucity of ground data on key human-related outcomes and the growing abundance and resolution of satellite imagery are quantified and reviewed, highlighting how this noise often leads to incorrect assessment of models’ predictive performance.
YOLO9000: Better, Faster, Stronger
YOLO9000, a state-of-the-art, real-time object detection system that can detect over 9000 object categories, is introduced and a method to jointly train on object detection and classification is proposed, both novel and drawn from prior work.