The iNaturalist Species Classification and Detection Dataset

Paper: Link

Objective:

  • Focuses on species of plants and animals captured in wide variety of situations, different camera types, varying image quality, feature large class imbalance and verified by citizen scientists.

Summary:

  • Details: There are a total of 5,089 categories in the dataset, with 579,184 training images and 95,986 validation images. For the training set, the distribution of images per category follows the observation frequency of that category by the iNaturalist community. Therefore, there is a non-uniform distribution of images per category.

  • Experiments: Classification experiments were done using ResNet, Inception V3, Inception ResNet V2 and MobileNet

  • Known issues: a.) Doesn’t contains additional annotations such as sex and life stage attributes, habitat tags, and pixel level labels for the four super-classes that were challenging to annotate. b.) Need of an efficient algorithm that works when the test set contains classes that were never seen during training.

Srishti Yadav
Srishti Yadav
ML Researcher

My research interest include applying computationally intensive machine learning algorithm to computer vision algoritmns.

Related