The iNaturalist Species Classification and Detection Dataset-Supplementary Material

By Grant Van Horn, Oisin Mac Aodha, Yang Song, Yin Cui, Chenghe Sun, Alexander Shepard, Hartwig Adam. 2018

Subfields: Computer Vision and Pattern Recognition, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Jiebo Luo
2. Zhengjun Zha
3. Jianlong Fu
4. Luc Van Gool
5. Andrew Zisserman
6. Dacheng Tao
7. Ishan Misra
8. Trevor Darrell

In Evaluation & Benchmarks

1. Zhengjun Zha
2. L. V. D. van der Maaten
3. Ishan Misra
4. Heliang Zheng
5. Jianlong Fu
6. Jiebo Luo

Citations in the corpus

• Masked Autoencoders Are Scalable Vision Learners (2021, neutral): “Table 6 studies transfer learning on the iNaturalists [56] and Places [71] tasks (see A.”
• Momentum Contrast for Unsupervised Visual Representation Learning (2019, neutral): “Besides, MoCo is on par on Cityscapes instance segmentation, and lags behind on VOC semantic segmentation; we show another comparable case on iNaturalist [57] in appendix.”
• Self-Supervised Learning of Pretext-Invariant Representations (2019, methodology): “We follow the evaluation setup from [19] and measure the performance of such classifiers on four image-classification datasets: ImageNet [58], VOC07 [15], Places205 [79], and iNaturalist2018 [65].”
• Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation (2020, neutral): “Recently, the computer vision community has begun to focus on the long-tail nature of object categories present in natural images [59, 21], where many of the different object categories have very few labeled images.”
• Hiera: A Hierarchical Vision Transformer without the Bells-and-Whistles (2023, unknown): “Notation: setting corresponding to 800 / 1600 / 3200 epochs of pretraining. uralist2019 (Van Horn et al., 2018) and Places365 (Zhou et al., 2014).”
• Exponential Moving Average Normalization for Self-supervised and Semi-supervised Learning (2021, unknown): “Pascal VOC2007 [13] and iNaturalist [23] have also been tested.”
• Omnivore: A Single Model for Many Visual Modalities (2022, unknown): “We evaluate OMNIVORE on fine-grained object recognition on the iNaturalist-2018 dataset [36], finegrained classification on the Oxford-IIIT Pets dataset [69], and in scene classification on the Places-365 dataset [105].”
• Looking for the Devil in the Details: Learning Trilinear Attention Sampling Network for Fine-Grained Image Recognition (2019, methodology): “CUB-200-2011 [34] 200 5,994 5,794 Stanford-Car [13] 196 8,144 8,041 iNaturalist-2017 [27] 5,089 579,184 95,986”
• Rethinking the Hyperparameters for Fine-tuning (2020, unknown): “256 15,360 15,189 iNaturalist 2017 (Van Horn et al., 2018) fine-grained object recog.”
• Learning Deep Bilinear Transformation for Fine-grained Image Representation (2019, unknown): “Large-scale image recognition benchmarks: To further evaluate the proposed DBTNet on large scale image datasets, we conduct experiments on iNaturalist-2017 [34].”