diff --git a/README.md b/README.md index 4733b77..4f94b03 100644 --- a/README.md +++ b/README.md @@ -1,4 +1,5 @@ -# Citrus-Farm-Dataset +![Leading Image](docs/figures/leading_image.png) + ## Introduction CitrusFarm is a multimodal agricultural robotics dataset that provides both **multispectral images** and **navigational sensor data** for localization, mapping and crop monitoring tasks. - It was collected by a wheeled mobile robot in the [Agricultural Experimental Station](https://cnas.ucr.edu/resources/agricultural-experiment-station) at the [University of California Riverside](https://www.ucr.edu/) in the summer of 2023. @@ -8,6 +9,8 @@ CitrusFarm is a multimodal agricultural robotics dataset that provides both **mu **Authors:** Hanzhe Teng, Yipeng Wang, Xiaoao Song and Konstantinos Karydis from [ARCS Lab](https://sites.google.com/view/arcs-lab/) at [UC Riverside](https://www.ucr.edu/). +**Related Workshop Spotlight:** [Present and Future of Agricultural Robotics and Technologies: Academic and Industry Perspectives](https://sites.google.com/view/agrobotics) (IROS 2023) + **Related Publications:** H. Teng, Y. Wang, X. Song and K. Karydis, "Multimodal Dataset for Localization, Mapping and Crop Monitoring in Citrus Tree Farms", in the 18th International Symposium on Visual Computing (ISVC 2023). ([preprint](https://arxiv.org/abs/2309.15332)) ``` diff --git a/docs/calibration.md b/docs/calibration.md index 70ec568..ae93e4d 100644 --- a/docs/calibration.md +++ b/docs/calibration.md @@ -20,7 +20,7 @@ It is worth noting that `base_link` is the frame used by wheel odometry, and is Other notes regarding GPS frame: - With one GPS receiver on the robot, we can access only the 3D position (rather than full 6-DoF pose) of the robot. Therefore, the orientation (quaternion) component of LiDAR-GPS extrinsic parameters is not very meaningful. -- More precisely, the GPS-RTK data is with respect to the center of the GPS receiver; if your algorithm output is expressed in IMU or LiDAR frame, it is better to convert it to the GPS frame first and align orientation using SVD, before computing the actual error (e.g., Absolute Trajectory Error (ATE)). We provide such evaluation scripts in the [tools](tools.html) as well. +- More precisely, the GPS-RTK data is with respect to the center of the GPS receiver; if your algorithm output is expressed in IMU or LiDAR frame, it is better to first convert it to the GPS frame and align orientation using SVD, before computing the actual error (e.g., Absolute Trajectory Error (ATE)). We provide such evaluation scripts in the [tools](tools.html) as well. ### Calibration Process ![Calibration images](./figures/calibration_image.png) diff --git a/docs/figures/leading_image.png b/docs/figures/leading_image.png new file mode 100644 index 0000000..9de3cdf Binary files /dev/null and b/docs/figures/leading_image.png differ diff --git a/docs/figures/logos/IROS.jpg b/docs/figures/logos/IROS.jpg new file mode 100644 index 0000000..a53358c Binary files /dev/null and b/docs/figures/logos/IROS.jpg differ diff --git a/docs/figures/image.png b/docs/figures/multispectral.png similarity index 100% rename from docs/figures/image.png rename to docs/figures/multispectral.png diff --git a/docs/index.md b/docs/index.md index eab921b..e869546 100644 --- a/docs/index.md +++ b/docs/index.md @@ -1,9 +1,11 @@ --- layout: article -title: CitrusFarm Dataset +# title: CitrusFarm Dataset --- -## Introduction +![Leading Image](./figures/leading_image.png) + +## CitrusFarm Dataset CitrusFarm is a multimodal agricultural robotics dataset that provides both **multispectral images** and **navigational sensor data** for localization, mapping and crop monitoring tasks. - It was collected by a wheeled mobile robot in the [Agricultural Experimental Station](https://cnas.ucr.edu/resources/agricultural-experiment-station) at the [University of California Riverside](https://www.ucr.edu/) in the summer of 2023. - It offers a total of **nine** sensing modalities, including stereo RGB, depth, monochrome, near-infrared and thermal images, as well as wheel odometry, LiDAR, IMU and GPS-RTK data. @@ -12,6 +14,8 @@ CitrusFarm is a multimodal agricultural robotics dataset that provides both **mu **Authors:** Hanzhe Teng, Yipeng Wang, Xiaoao Song and Konstantinos Karydis from [ARCS Lab](https://sites.google.com/view/arcs-lab/) at [UC Riverside](https://www.ucr.edu/). +**Related Workshop Spotlight:** [Present and Future of Agricultural Robotics and Technologies: Academic and Industry Perspectives](https://sites.google.com/view/agrobotics) (IROS 2023) + **Related Publications:** H. Teng, Y. Wang, X. Song and K. Karydis, "Multimodal Dataset for Localization, Mapping and Crop Monitoring in Citrus Tree Farms", in the 18th International Symposium on Visual Computing (ISVC 2023). ([preprint](https://arxiv.org/abs/2309.15332)) ``` @@ -31,8 +35,9 @@ This work is supported in part by NSF, USDA-NIFA, ONR and the University of Cali Furthermore, we thank Dr. Peggy Mauk and the staff team at UCR's Agricultural Experimental Station for their support in our work. NSF logo -ONR logo +ONR logo USDA logo LACA logo -ISVC2023 logo -ARCS logo +IROS logo +ISVC2023 logo +ARCS logo diff --git a/docs/sequence.md b/docs/sequence.md index 51c375d..c20a7af 100644 --- a/docs/sequence.md +++ b/docs/sequence.md @@ -19,4 +19,4 @@ title: Sequences ![Sample Trajectories](./figures/trajectory.png) ## Sample Multispectral Images -![Sample Multispectral Images](./figures/image.png) +![Sample Multispectral Images](./figures/multispectral.png)