This repository contains the team's work in PlantCLEF 2021. The goal of the challenge was to identify plants in field pictures based on a training set of digitized herbarium specimens. We achieved an MRR of 0.181 on the primary metric and an MRR of 0.158 on the secondary metric (difficult species). The official results are released here.
- HFTL Network
- run_HFTL_network.py (Main script)
- network_module_HFTL.py
- OSM Network
- run_OSM_network.py (Main script)
- network_module_OSM.py
-
HFTL Network
-
OSM Network
-
Herbarium Network
-
Field Network (2017)
-
Field Network (2021)
-
HFTL Network
- Herbarium Train
list/HFTL/clef2020_known_classes_herbarium_train_added.txt - Herbarium Test
list/HFTL/clef2020_known_classes_herbarium_test.txt - Field Train
list/HFTL/clef2020_known_classes_field_train_added.txt - Field Test
list/HFTL/clef2020_known_classes_field_test.txt
- Herbarium Train
-
OSM Network
-
Herbarium Dictionary
- Without field bias
list/997_class_emb_sample_for_herbDict_server_removed_duplicates.txt - With field bias
list/997_class_emb_sample_for_herbDict_new_field_bias.txt
- Without field bias
-
Test Sets
- Test Set 1 (With field training images)
list/HFTL/clef2020_known_classes_field_test.txt - Test Set 2 (Without field training images)
list/missing_class_sample.txt
(Note: The image files for Test Set 2 are sourced from Google Image queries)
- Test Set 1 (With field training images)
- HFTL Network
- OSM Network