ASRL // In the Dark and Multiseason

UTIAS Long-Term Localization and Mapping Dataset

Overview • Related Work • Hardware • Description • Tools • Datasets

Overview

The UTIAS In the Dark and Multiseason datasets contain 159,549 stereo image pairs collected at the University of Toronto Institute for Aerospace Studies. In both experiments, a closed loop path was manually driven to create a Teach run then the robot autonomously re-traversed the path at staggered time intervals creating Repeat runs. Stereo keyframes were captured approximately every 0.2 metres with all keyframes accurately localized back to the Teach run. Relative transforms for each stereo pair are provided in human-readable text files. A set of Python tools is provided to work with the data. As few or as many repeat runs as desired can be downloaded. Potential uses for these datasets include research in deep visual localization, deep visual odometry, all-weather localization, and experience-based localization. They may also be useful for work on dense stereo reconstruction, simultaneous localization and mapping, robust image mapping, or saliency detection. See Related Work for more past uses.

In the Dark

The UTIAS In the Dark dataset is a collection of 39 runs of a 260 metre path that were collected approximately once an hour over 30 hours. Headlights were used to continue traversal at night. This provides a full cycle of daily lighting conditions that may be used for comparison. Challenges include high sun glare as well as varying shadows occurring at dusk and dawn. The scenery is a mix of buildings, fencing, roads, and off-road vegetation. In this dataset, low-grade GPS data is provided as an extra for 23 of the 39 runs.

Multiseason

The UTIAS Multiseason dataset is a collection of 136 runs of a separate 160 metre loop that were collected approximately once per day from January 31, 2017 to May 27, 2017. The runs cover a range of seasonal appearances from mixed snow cover over dead vegetation to fully grown grass and shrubs in warm weather. They also feature a range of illumination changes due to both the varied start time of runs and the changing sun position throughout the spring. The scenery is mainly natural vegetation with some man-made structures. Weather varies from snowfall to freezing rain to overcast skies to full sun. High wind on later runs causes changes in vegetation geometry providing a challenge to visual odometry and localization algorithms.

Related Work

UTIAS In the Dark and Multiseason were collected using Visual Teach & Repeat (VT&R), a system to allow highly accurate autonomous traversal of a previously manually driven path using only a vision sensor. For more information on VT&R, see Furgale and Barfoot (2010). These datasets are unique in capturing stereo imagery of the same off-road loop many times in different conditions. Accurate ground-truth is provided by VT&R's Multi-Experience Localization (MEL). It uses bridging experiences to match images captured under lighting and weather conditions that are drastically different from the original Teach pass. See MacTavish et al. (2018) for more information.

These datasets are most similar to the Oxford RobotCar Dataset. However, ours feature off-road terrain and the accurate autonomous traversals of VT&R provide a higher degree of viewpoint invariance. UTIAS In the Dark also captures short-term illumination changes in a systematic way, unconfounded by long-term structural or seasonal change. They also share similarities with the Symphony Lake Dataset in capturing a variety of weather and seasonal changes through repeated surveys of a natural environment. The Symphony Lake Dataset does not capture stereo data, however, and the same route is only approximately followed each run. Like the New College Dataset, our dataset captures stereo imagery of the same loop of a mixed environment, though ours was collected over multiple runs providing a far greater amount of appearance change. We believe these datasets are a novel contribution to the robotics and computer vision communities and will be of use to many researchers.

The following publications have used these datasets in their work. If you use one or both of these datasets, please email ben.congram [at] robotics.utias.utoronto.ca to be added to the list.

[1] Gridseth M and Barfoot T D. “DeepMEL: Compiling Visual Multi-Experience Localization into a Deep Neural Network”. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA). Paris, France, 31 May - 4 June 2020.

[2] Clement L E, Gridseth M, Tomasi J, and Kelly J. "Learning Matchable Image Transformations for Long-term Metric Visual Localization". IEEE Robotics and Automation Letters (RAL), 5(2):1492–1499, January 2020. doi: 10.1109/LRA.2020.2967659.

[3] Gridseth M and Barfoot T D. “Towards Direct Localization for Visual Teach and Repeat”. In Proceedings of the 16th Conference on Computer and Robot Vision (CRV). Kingston, Canada, 29-31 May 2019. doi: 10.1109/CRV.2019.00021.

[4] Zhang N, Warren M, and Barfoot T D. “Learning Place-and-Time-Dependent Binary Descriptors for Long-Term Visual Localization”. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA). Brisbane, Australia, 21-25 May 2018. doi: 10.1109/ICRA.2018.8460674.

[5] MacTavish K A, Paton M, and Barfoot T D. “Selective Memory: Recalling Relevant Experience for Long-Term Visual Localization”. Journal of Field Robotics, 35(8):1265–1292, 2018. doi: 10.1002/rob.21838.

[6] MacTavish K A, Paton M, and Barfoot T D. “Night Rider: Visual Odometry Using Headlights”. In Proceedings of the 14th Conference on Computer and Robot Vision (CRV), pages 314–320. Edmonton, Alberta, 16-19 May 2017. doi: DOI 10.1109/CRV.2017.48.

[7] MacTavish K A, Paton M, and Barfoot T D. “Visual Triage: A Bag-of-Words Experience Selector for Long-Term Visual Route Following”. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pages 2065–2072. Singapore, 29 May - 3 June 2017. doi: 10.1109/ICRA.2017.7989238.

[8] Paton M, MacTavish K A, Warren M, and Barfoot T D. “Bridging the Appearance Gap: Multi-Experience Localization for Long-Term Visual Teach and Repeat”. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 1918–1925. Daejeon, Korea, 9-14 October 2016. doi: 10.1109/IROS.2016.7759303.

Hardware

Both datasets were captured aboard a Clearpath Robotics Grizzly Robotic Utility Vehicle. A Point Grey Bumblebee XB3 stereo camera was rigidly mounted to a mast on the Grizzly. The XB3 provides a 0.24 metre baseline and 512 x 384 resolution. The full stereo stream was collected at 16 Hz but has been downsampled with stereo keyframes provided approximately every 0.2 metres travelled. On-board headlights were rigidly mounted below the stereo camera and used to illuminate scenery during the night-time traversals of the In the Dark dataset. A GPS unit logged low-grade GPS data for a portion of the In the Dark dataset runs (though was not used to assist the autonomous traversals).

The camera parameters are given here: camera_parameters.txt

The transformation matrix describing the rigid transform from the vehicle frame to the camera frame, T_camera_vehicle, is given here: transform_camera_vehicle.txt

The transformation matrix describing the rigid transform from the vehicle frame to the GPS frame, T_gps_vehicle, is given here: transform_gps_vehicle.txt

Description of Data Products

Pose Graph

To work with a dataset, start by downloading the Teach run, run_000000, and as many repeat runs as desired. Keyframes in the Teach run have been localized to the following Teach run keyframe via visual odometry. All repeat run keyframes are localized back to the Teach run via visual localization. The relative transformation between any two vertices in the dataset can be found by chaining poses through the Teach run. Note: transforms between topologically near vertices in the pose graph can be considered very accurate, but transformations between distant vertices should not be relied upon. The figure below illustrates the pose graph form with dots representing vertices and arrows representing transformations. The first number in the vertex ID tuple indicates the run number while the second number indicates the pose within the run.

Coordinate Frames

The figure to the right illustrates the coordinate frames applicable to these datasets.

All transformation data is given as relative transformations of the vehicle frame between vertices. The vehicle frame is approximately situated in the centre of the Grizzly at axle level. Images are captured in the camera frame. The camera frame is centered at the left stereo camera. GPS measurements are recorded as absolute measurements of the GPS frame with respect to an inertial frame. The 4x4 homogeneous transformation matrices describing the rigid transformations between the sensor and vehicle frames are provided in the Hardware section.

Transformation matrices use the T_to_from convention. For example, to transform a point expressed in frame_a to its coordinates expressed in frame_b one would do the following:

Here, p_jb_in_b is the coordinates of point j with respect to point b in frame_b , and p_ja_in_a is the coordinates of point j with respect to point a in frame_a .

C_ba is the rotation matrix from frame_a to frame_b and r_ab is the translation vector from to , expressed in .

File Structure

The datasets are split into run folders. run_000017 contains the 17^th repeat run. In each run folder, there is an image folder with subdirectories for the left and right stereo images. Images are named after their corresponding vertices. For example, run_000004/images/left/000065.png corresponds to vertex (4,65). Also in the run folder is a timestamps_images.txt file and a transformation file — either transforms_temporal.txt for Teach runs, or transforms_spatial.txt for Repeat runs. GPS measurements, when available, are provided in a separate text file, gps.txt, and their timestamps in timestamps_gps.txt.

Interpreting Text Files

Transformation Files

The transformation data is contained in human-readable, comma-delimited text files. Each row provides the SE(3) transformation matrix between two vertices. The first two values represent the vertex ID of vertex A, the second two values the ID of vertex B, then the remaining 16 values provide the transformation matrix in row-major order.

Timestamp Files

The timestamp files are also human-readable and comma-delimited with each row representing the timestamp of one vertex. The first value is the vertex number and the second a Unix timestamp in nanoseconds. These may be converted to UTC via standard datetime library functions.

GPS Files

The GPS measurement files are human-readable and comma-delimited with each row representing the closest GPS measurement to a particular vertex. The run number and vertex number are listed first followed by latitude, longitude, and altitude. A file containing the Unix timestamps of these measurements is also included.

Helpful Tools

To simplify processing this data, Python tools have been provided in a GitHub repository. The scripts parse the text files to build a pose graph from the stored runs. The created graph can be used to calculate both the metric transformations and topological distance between particular vertices. It can also be queried to return other vertices within a particular radius of a given vertex.

Dataset

This section provides individual zip files for each run containing the stereo images and text files for the transformations, timestamps, and GPS measurements if available. A Python downloader script has also been provided to aid in downloading large portions of the datasets. A video preview is provided for each run to quickly see weather and lighting conditions before downloading. Note: some runs in the Multiseason dataset do not form a complete loop.

Important note: This dataset provides ftp download links. Some browsers no longer enable ftp links by default and you need to change the settings if you want to download using the browser. Using the wget command or the download script is an alternative to avoid the browser.

Skip to Multiseason

In the Dark

Run	Start Time	Description	GPS	Number of Frames	Download
000000	July 19 - 08:58 EDT	morning, sunny	Yes	1116	.zip [905 MB]
000001	July 19 - 09:09 EDT	morning, sunny, small shadows	Yes	1195	.zip [924 MB]
000002	July 19 - 09:46 EDT	morning, sunny, small shadows	Yes	1092	.zip [837 MB]
000003	July 19 - 10:58 EDT	morning, sunny, small shadows	No	1122	.zip [854 MB]
000004	July 19 - 12:29 EDT	mid-day, cloudy	No	981	.zip [722 MB]
000005	July 19 - 16:48 EDT	afternoon, sunny, small shadows	Yes	1086	.zip [818 MB]
000006	July 19 - 18:36 EDT	evening, sunny, medium shadows, large sun glare	Yes	1602	.zip [1117 MB]
000007	July 19 - 19:31 EDT	evening, sunny, long shadows, large sun glare	No	2220	.zip [1393 MB]
000008	July 19 - 20:01 EDT	evening, low sun, large sun glare	No	2652	.zip [1485 MB]
000009	July 19 - 20:17 EDT	evening, low sun, large sun glare	Yes	2505	.zip [1407 MB]
000010	July 19 - 20:30 EDT	evening, low sun, small sun glare	Yes	958	.zip [651 MB]
000011	July 19 - 20:44 EDT	dusk, low sun	Yes	886	.zip [607 MB]
000012	July 19 - 21:00 EDT	dusk, low sun	Yes	862	.zip [590 MB]
000013	July 19 - 21:12 EDT	dusk, little natural light	Yes	2226	.zip [1613 MB]
000014	July 19 - 21:26 EDT	dusk, little natural light	Yes	3335	.zip [2368 MB]
000015	July 19 - 21:37 EDT	dusk, little natural light	Yes	3710	.zip [2611 MB]
000016	July 19 - 21:48 EDT	dusk, dark	Yes	3634	.zip [2537 MB]
000017	July 19 - 21:58 EDT	nighttime, dark	Yes	3907	.zip [2723 MB]
000018	July 19 - 22:28 EDT	nighttime, dark	Yes	3701	.zip [2586 MB]
000019	July 19 - 22:40 EDT	nighttime, dark	Yes	3718	.zip [2597 MB]
000020	July 20 - 04:29 EDT	nighttime, dark	No	4042	.zip [2837 MB]
000021	July 20 - 05:10 EDT	dawn, little natual light	No	3754	.zip [2679 MB]
000022	July 20 - 05:23 EDT	dawn, little natual light	No	2761	.zip [1941 MB]
000023	July 20 - 05:34 EDT	dawn, light sky	No	1274	.zip [881 MB]
000024	July 20 - 05:48 EDT	dawn, light sky	No	964	.zip [671 MB]
000025	July 20 - 06:04 EDT	dawn, light sky	Yes	1012	.zip [711 MB]
000026	July 20 - 06:16 EDT	dawn, light sky	Yes	1111	.zip [763 MB]
000027	July 20 - 06:31 EDT	dawn, large sun glare	Yes	2613	.zip [1565 MB]
000028	July 20 - 06:44 EDT	dawn, large sun glare	Yes	1826	.zip [1187 MB]
000029	July 20 - 07:03 EDT	dawn, large sun glare	Yes	1873	.zip [1184 MB]
000030	July 20 - 07:19 EDT	dawn, medium shadows, large sun glare	Yes	1716	.zip [1153 MB]
000031	July 20 - 07:39 EDT	dawn, medium shadows, small sun glare	Yes	1173	.zip [841 MB]
000032	July 20 - 08:35 EDT	morning, medium shadows, small sun glare	No	1034	.zip [772 MB]
000033	July 20 - 09:18 EDT	morning, sunny, small shadows	No	1045	.zip [798 MB]
000034	July 20 - 10:21 EDT	morning, cloudy, small shadows	No	986	.zip [736 MB]
000035	July 20 - 11:26 EDT	mid-day, sunny, small shadows	No	1080	.zip [822 MB]
000036	July 20 - 12:29 EDT	mid-day, cloudy, small shadows	No	1071	.zip [815 MB]
000037	July 20 - 14:01 EDT	afternoon, cloudy, small shadows	No	989	.zip [752 MB]
000038	July 20 - 15:30 EDT	afternoon, cloudy, small shadows	No	950	.zip [724 MB]

Multiseason

Run	Start Time	Description	Number of Frames	Download
000000	January 31 - 12:39 EST	mid-day, overcast, little snow, brown grass	826	.zip [592 MB]
000001	January 31 - 12:48 EST	mid-day, overcast, little snow, brown grass	827	.zip [593 MB]
000002	January 31 - 12:55 EST	mid-day, overcast, little snow, brown grass	849	.zip [612 MB]
000003	January 31 - 15:04 EST	afternoon, overcast, little snow, brown grass	716	.zip [507 MB]
000004	February 01 - 12:23 EST	mid-day, cloudy, some snow, brown grass	717	.zip [476 MB]
000005	February 02 - 15:32 EST	afternoon, overcast, some snow, brown grass	702	.zip [544 MB]
000006	February 02 - 17:19 EST	dusk, overcast, some snow, brown grass	631	.zip [478 MB]
000007	February 03 - 11:18 EST	morning, sunny, some snow, brown grass	757	.zip [611 MB]
000008	February 03 - 12:57 EST	mid-day, cloudy, some snow, brown grass	721	.zip [569 MB]
000009	February 03 - 13:13 EST	mid-day, cloudy, some snow, brown grass	686	.zip [539 MB]
000010	February 03 - 14:38 EST	afternoon, cloudy, some snow, brown grass	844	.zip [697 MB]
000011	February 03 - 15:16 EST	afternoon, cloudy, some snow, brown grass	769	.zip [618 MB]
000012	February 07 - 08:59 EST	morning, overcast, little snow, brown grass	933	.zip [716 MB]
000013	February 07 - 09:04 EST	morning, overcast, little snow, brown grass	872	.zip [670 MB]
000014	February 07 - 14:16 EST	afternoon, overcast, little snow, brown grass	1218	.zip [772 MB]
000015	February 07 - 15:50 EST	afternoon, cloudy, little snow, brown grass	1104	.zip [704 MB]
000016	February 07 - 16:22 EST	afternoon, cloudy, little snow, brown grass	1315	.zip [834 MB]
000017	February 07 - 17:17 EST	dusk, cloudy, little snow, brown grass	1138	.zip [702 MB]
000018	February 08 - 17:20 EST	dusk, cloudy, little snow, brown grass	725	.zip [557 MB]
000019	February 08 - 17:41 EST	dusk, cloudy, snowing lightly, little snow, brown grass	850	.zip [642 MB]
000020	February 08 - 17:54 EST	dusk, cloudy, snowing lightly, little snow, brown grass	1640	.zip [1241 MB]
000021	February 08 - 18:22 EST	nighttime, cloudy, snowing lightly, little snow, brown grass	652	.zip [472 MB]
000022	February 09 - 15:06 EST	afternoon, cloudy, little snow, brown grass	579	.zip [472 MB]
000023	February 09 - 16:34 EST	afternoon, sunny, little snow, brown grass	579	.zip [476 MB]
000024	February 10 - 12:15 EST	mid-day, cloudy, little snow, brown grass	576	.zip [456 MB]
000025	February 10 - 12:22 EST	mid-day, cloudy, little snow, brown grass	575	.zip [454 MB]
000026	February 10 - 12:30 EST	mid-day, cloudy, little snow, brown grass	537	.zip [423 MB]
000027	February 10 - 14:54 EST	afternoon, overcast, some snow, brown grass	592	.zip [417 MB]
000028	February 10 - 15:06 EST	afternoon, overcast, some snow, brown grass	595	.zip [422 MB]
000029	February 10 - 16:08 EST	afternoon, overcast, some snow, brown grass	638	.zip [429 MB]
000030	February 10 - 16:20 EST	afternoon, overcast, some snow, brown grass	730	.zip [492 MB]
000031	February 10 - 17:25 EST	dusk, overcast, some snow, brown grass	687	.zip [426 MB]
000032	February 11 - 12:03 EST	mid-day, overcast, some snow, brown grass	641	.zip [442 MB]
000033	February 11 - 16:18 EST	afternoon, overcast, some snow, brown grass	588	.zip [433 MB]
000034	February 13 - 14:38 EST	afternoon, overcast, full snow	714	.zip [481 MB]
000035	February 13 - 15:45 EST	afternoon, cloudy, full snow	636	.zip [449 MB]
000036	February 13 - 16:55 EST	afternoon, cloudy, full snow, small sun glare	652	.zip [466 MB]
000037	February 13 - 17:09 EST	afternoon, cloudy, full snow, small sun glare	749	.zip [540 MB]
000038	February 13 - 17:47 EST	dusk, cloudy, full snow	650	.zip [448 MB]
000039	February 13 - 18:02 EST	dusk, cloudy, full snow	605	.zip [387 MB]
000040	February 13 - 18:11 EST	dusk, cloudy, full snow	640	.zip [367 MB]
000041	February 13 - 18:31 EST	nighttime, cloudy, full snow	702	.zip [384 MB]
000042	February 14 - 11:38 EST	mid-day, cloudy, full snow	650	.zip [370 MB]
000043	February 14 - 12:17 EST	mid-day, cloudy, full snow	845	.zip [338 MB]
000044	February 14 - 13:33 EST	mid-day, cloudy, full snow	611	.zip [367 MB]
000045	February 14 - 14:01 EST	afternoon, cloudy, full snow	598	.zip [368 MB]
000046	February 15 - 14:03 EST	afternoon, cloudy, full snow	577	.zip [361 MB]
000047	February 15 - 15:20 EST	afternoon, overcast, some snow, brown grass	589	.zip [424 MB]
000048	February 16 - 11:46 EST	mid-day, sunny, some snow, brown grass, partial camera obstruction	625	.zip [311 MB]
000049	February 16 - 14:23 EST	afternoon, sunny, some snow, brown grass	577	.zip [366 MB]
000050	February 16 - 15:11 EST	afternoon, sunny, some snow, brown grass	571	.zip [411 MB]
000051	February 16 - 17:11 EST	dusk, sunny, some snow, brown grass, medium sun glare	615	.zip [455 MB]
000052	February 17 - 13:43 EST	mid-day, sunny, some snow, brown grass	575	.zip [354 MB]
000053	February 18 - 13:38 EST	mid-day, sunny, some snow, brown grass, muddy	580	.zip [450 MB]
000054	February 18 - 15:12 EST	afternoon, cloudy, little snow, brown grass, muddy	582	.zip [480 MB]
000055	February 18 - 16:08 EST	afternoon, sunny, little snow, brown grass, muddy	573	.zip [474 MB]
000056	February 18 - 17:32 EST	dusk, cloudy, little snow, brown grass, muddy	581	.zip [422 MB]
000057	February 19 - 13:12 EST	mid-day, sunny, little snow, brown grass, muddy	588	.zip [509 MB]
000058	February 19 - 15:13 EST	afternoon, sunny, little snow, brown grass, muddy	600	.zip [525 MB]
000059	February 19 - 16:14 EST	afternoon, sunny, brown grass, small sun glare, muddy	614	.zip [516 MB]
000060	February 21 - 13:18 EST	mid-day, overcast, brown grass, muddy	592	.zip [492 MB]
000061	February 21 - 13:25 EST	mid-day, overcast, brown grass, muddy	591	.zip [491 MB]
000062	February 22 - 13:00 EST	mid-day, overcast, brown grass, muddy	592	.zip [491 MB]
000063	February 24 - 14:58 EST	afternoon, sunny, brown grass, muddy	588	.zip [494 MB]
000064	February 27 - 14:10 EST	afternoon, sunny, brown grass	594	.zip [525 MB]
000065	February 27 - 18:04 EST	dusk, overcast, brown grass	582	.zip [432 MB]
000066	February 28 - 09:49 EST	morning, overcast, brown grass	583	.zip [488 MB]
000067	February 28 - 11:34 EST	mid-day, overcast, brown grass	568	.zip [478 MB]
000068	March 03 - 12:23 EST	mid-day, cloudy, brown grass	588	.zip [514 MB]
000069	March 03 - 12:28 EST	mid-day, cloudy, brown grass	581	.zip [503 MB]
000070	March 07 - 13:53 EST	mid-day, overcast, brown grass, muddy	578	.zip [454 MB]
000071	March 08 - 13:39 EST	mid-day, overcast, brown grass	583	.zip [474 MB]
000072	March 08 - 15:14 EST	afternoon, sunny, brown grass, small sun glare	577	.zip [497 MB]
000073	March 10 - 13:07 EST	mid-day, cloudy, brown grass	584	.zip [504 MB]
000074	March 13 - 12:54 EDT	mid-day, overcast, snowing lightly, little snow, brown grass	529	.zip [443 MB]
000075	March 13 - 14:29 EDT	afternoon, overcast, snowing lightly, some snow, brown grass	571	.zip [453 MB]
000076	March 13 - 15:35 EDT	afternoon, overcast, some snow, brown grass	569	.zip [456 MB]
000077	March 13 - 17:35 EDT	afternoon, overcast, some snow, brown grass	572	.zip [466 MB]
000078	March 14 - 15:23 EDT	afternoon, overcast, some snow, brown grass	575	.zip [408 MB]
000079	March 14 - 18:53 EDT	dusk, overcast, some snow, brown grass	578	.zip [419 MB]
000080	March 15 - 18:52 EDT	dusk, sunny, some snow, brown grass, small sun glare	567	.zip [443 MB]
000081	March 17 - 11:02 EDT	morning, sunny, little snow, brown grass, muddy	564	.zip [468 MB]
000082	March 17 - 11:12 EDT	morning, sunny, little snow, brown grass, muddy	590	.zip [494 MB]
000083	March 17 - 15:12 EDT	afternoon, cloudy, little snow, brown grass, muddy	583	.zip [492 MB]
000084	March 17 - 17:27 EDT	afternoon, cloudy, little snow, brown grass, muddy	566	.zip [459 MB]
000085	March 21 - 15:18 EDT	afternoon, sunny, brown grass	594	.zip [525 MB]
000086	March 21 - 15:27 EDT	afternoon, sunny, brown grass	588	.zip [518 MB]
000087	March 21 - 19:40 EDT	dusk, sunny, brown grass	591	.zip [428 MB]
000088	March 23 - 12:36 EDT	mid-day, sunny, brown grass	591	.zip [518 MB]
000089	March 27 - 16:10 EDT	afternoon, overcast, brown grass	587	.zip [489 MB]
000090	March 27 - 16:18 EDT	afternoon, overcast, brown grass	587	.zip [491 MB]
000091	March 27 - 16:23 EDT	afternoon, overcast, brown grass	589	.zip [483 MB]
000092	March 27 - 16:28 EDT	afternoon, overcast, brown grass	588	.zip [485 MB]
000093	March 29 - 13:55 EDT	mid-day, overcast, brown grass	592	.zip [499 MB]
000094	March 29 - 14:01 EDT	mid-day, overcast, brown grass	587	.zip [489 MB]
000095	March 31 - 09:44 EDT	morning, overcast, some snow, brown grass	578	.zip [444 MB]
000096	March 31 - 11:41 EDT	mid-day, overcast, little snow, brown grass, muddy	557	.zip [420 MB]
000097	April 03 - 13:29 EDT	mid-day, overcast brown grass	525	.zip [443 MB]
000098	April 04 - 13:42 EDT	mid-day, overcast brown grass, muddy	536	.zip [423 MB]
000099	April 06 - 17:49 EDT	afternoon, raining, green grass, muddy	590	.zip [468 MB]
000100	April 07 - 12:21 EDT	mid-day, overcast, some snow, muddy	596	.zip [455 MB]
000101	April 07 - 14:01 EDT	mid-day, cloudy, some snow, green grass muddy	576	.zip [475 MB]
000102	April 07 - 14:07 EDT	mid-day, cloudy, some snow, green grass muddy	578	.zip [486 MB]
000103	April 10 - 16:19 EDT	afternoon, overcast, green grass, muddy	579	.zip [469 MB]
000104	April 10 - 17:43 EDT	afternoon, sunny, green grass, muddy, small sun glare	586	.zip [493 MB]
000105	April 10 - 17:49 EDT	afternoon, sunny, green grass, muddy, small sun glare	587	.zip [495 MB]
000106	April 11 - 18:36 EDT	dusk, cloudy, green grass, muddy	571	.zip [452 MB]
000107	April 13 - 16:42 EDT	afternoon, sunny, green grass, small sun glare	580	.zip [505 MB]
000108	April 13 - 16:47 EDT	afternoon, sunny, green grass, small sun glare	576	.zip [497 MB]
000109	April 17 - 16:52 EDT	afternoon, sunny, green grass, small sun glare	579	.zip [516 MB]
000110	April 17 - 19:53 EDT	dusk, sunny, green grass	573	.zip [429 MB]
000111	April 18 - 14:04 EDT	afternoon, sunny, green grass	571	.zip [514 MB]
000112	April 18 - 14:09 EDT	afternoon, sunny, green grass	570	.zip [513 MB]
000113	April 18 - 16:59 EDT	afternoon, sunny, green grass	577	.zip [511 MB]
000114	April 18 - 17:12 EDT	afternoon, sunny, green grass, small sun glare	573	.zip [504 MB]
000115	April 18 - 17:19 EDT	afternoon, sunny, green grass, small sun glare	578	.zip [506 MB]
000116	April 20 - 15:12 EDT	afternoon, overcast, green grass	576	.zip [476 MB]
000117	April 21 - 13:28 EDT	mid-day, overcast, green grass	567	.zip [472 MB]
000118	April 21 - 13:34 EDT	mid-day, overcast, green grass	568	.zip [475 MB]
000119	April 24 - 17:39 EDT	afternoon, cloudy, green grass, small sun glare	575	.zip [487 MB]
000120	April 25 - 15:35 EDT	afternoon, cloudy, green grass	563	.zip [464 MB]
000121	April 25 - 15:40 EDT	afternoon, cloudy, green grass	560	.zip [461 MB]
000122	April 27 - 11:24 EDT	mid-day, sunny, green grass	568	.zip [498 MB]
000123	April 27 - 11:29 EDT	mid-day, sunny, green grass	560	.zip [490 MB]
000124	May 03 - 14:10 EDT	afternoon, sunny, green grass, tall grass	576	.zip [515 MB]
000125	May 03 - 15:44 EDT	afternoon, sunny, green grass, tall grass	576	.zip [515 MB]
000126	May 03 - 18:31 EDT	afternoon, sunny, green grass, tall grass, small sun glare	587	.zip [489 MB]
000127	May 04 - 11:08 EDT	morning, cloudy, green grass, tall grass	570	.zip [469 MB]
000128	May 05 - 18:06 EDT	afternoon, overcast, green grass, tall grass, muddy	569	.zip [448 MB]
000129	May 08 - 13:03 EDT	mid-day, sunny, green grass, tall grass, dandelions	586	.zip [522 MB]
000130	May 08 - 13:08 EDT	mid-day, sunny, green grass, tall grass, dandelions	577	.zip [515 MB]
000131	May 08 - 17:27 EDT	afternoon, cloudy, green grass, tall grass, dandelions	560	.zip [460 MB]
000132	May 09 - 13:14 EDT	mid-day, cloudy, green grass, tall grass, dandelions	576	.zip [496 MB]
000133	May 09 - 13:18 EDT	mid-day, cloudy, green grass, tall grass, dandelions	569	.zip [471 MB]
000134	May 11 - 14:03 EDT	afternoon, cloudy, green grass, tall grass, dandelions	559	.zip [460 MB]
000135	May 11 - 15:12 EDT	afternoon, cloudy, green grass, tall grass, dandelions	558	.zip [459 MB]

License

MIT License

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.