andyvhuynh commited on
Commit
8e81f27
1 Parent(s): bf6b625

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +7 -7
README.md CHANGED
@@ -61,15 +61,15 @@ dataset_info:
61
  sequence: int64
62
  ---
63
 
64
- ### Nature Multi-View (NMV) Dataset Datacard
 
65
 
66
-
67
- # Dataset Overview
68
  - Total Images: Over 3 million ground-level and aerial image pairs
69
  - Species Covered: Over 6,000 native and introduced plant species
70
  - Geographic Focus: California, USA
71
 
72
- # Data Description
73
  - Ground-Level Images:
74
  - Sourced from iNaturalist open data on AWS.
75
  - Filters applied:
@@ -85,7 +85,7 @@ dataset_info:
85
  - RGB-Infrared images, 256x256 pixels, 60 cm-per-pixel resolution.
86
  - Centered on the latitude and longitude of the iNaturalist observation.
87
 
88
- # Dataset Splits
89
  - Training Set:
90
  - Full Training Set: 1,755,602 observations, 3,307,025 images
91
  - Labeled Training Sets:
@@ -98,11 +98,11 @@ dataset_info:
98
  - Test Set:
99
  - 182,618 observations, 334,887 images
100
 
101
- # Characteristics and Challenges
102
  - Long-Tail Distribution: The dataset exhibits a long-tail distribution common in natural world settings, making it a realistic benchmark for machine learning applications.
103
  - Geographic Bias: Reflects the geographic bias of citizen science data, with more observations from densely populated and visited regions like urban areas and National Parks.
104
  - Many-to-One Pairing: Multiple ground-level images are paired to the same aerial image, which may require special handling in machine learning models.
105
 
106
- # References
107
  - iNaturalist: www.inaturalist.org
108
  - nited States Department of Agriculture: NAIP Imagery.
 
61
  sequence: int64
62
  ---
63
 
64
+ # Nature Multi-View (NMV) Dataset Datacard
65
+ To encourage development of better machine learning methods for operating with diverse, unlabeled natural world imagery, we introduce Nature Multi-View (NMV), a multi-view dataset of over 3 million ground-level and aerial image pairs from over 1.75 million citizen science observations for over 6,000 native and introduced plant species across California.
66
 
67
+ ## Dataset Overview
 
68
  - Total Images: Over 3 million ground-level and aerial image pairs
69
  - Species Covered: Over 6,000 native and introduced plant species
70
  - Geographic Focus: California, USA
71
 
72
+ ## Data Description
73
  - Ground-Level Images:
74
  - Sourced from iNaturalist open data on AWS.
75
  - Filters applied:
 
85
  - RGB-Infrared images, 256x256 pixels, 60 cm-per-pixel resolution.
86
  - Centered on the latitude and longitude of the iNaturalist observation.
87
 
88
+ ## Dataset Splits
89
  - Training Set:
90
  - Full Training Set: 1,755,602 observations, 3,307,025 images
91
  - Labeled Training Sets:
 
98
  - Test Set:
99
  - 182,618 observations, 334,887 images
100
 
101
+ ## Characteristics and Challenges
102
  - Long-Tail Distribution: The dataset exhibits a long-tail distribution common in natural world settings, making it a realistic benchmark for machine learning applications.
103
  - Geographic Bias: Reflects the geographic bias of citizen science data, with more observations from densely populated and visited regions like urban areas and National Parks.
104
  - Many-to-One Pairing: Multiple ground-level images are paired to the same aerial image, which may require special handling in machine learning models.
105
 
106
+ ## References
107
  - iNaturalist: www.inaturalist.org
108
  - nited States Department of Agriculture: NAIP Imagery.