Transfer Learning with Self-Supervised Vision Transformers for Snake Identification

Anthony Miyaguchi, Murilo Gustineli, Austin Fischer, Ryan Lundqvist
7/8/2024
cs.CVcs.IRcs.LG

No official code found. Best implementation: dsgt-arc/snakeclef-2024 (4 stars, last commit 2024-07-02).

Abstract

We present our approach for the SnakeCLEF 2024 competition to predict snake species from images. We explore and use Meta's DINOv2 vision transformer model for feature extraction to tackle species' high variability and visual similarity in a dataset of 182,261 images. We perform exploratory analysis on embeddings to understand their structure, and train a linear classifier on the embeddings to predict species. Despite achieving a score of 39.69, our results show promise for DINOv2 embeddings in snake identification. All code for this project is available at https://github.com/dsgt-kaggle-clef/snakeclef-2024.

View on arXivView PDF

Code Implementations(1)

dsgt-arc/snakeclef-2024Community100%
41Shell, PythonJan 6, 20242 years ago

Similar Papers

Multi-Label Plant Species Classification with Self-Supervised Vision Transformers

83% match1 repo172024

CNN Based Flank Predictor for Quadruped Animal Species

79% match2024

Multi-Label Plant Species Prediction with Metadata-Enhanced Multi-Head Vision Transformers

78% match1 repo12025

Self-supervised Learning on Camera Trap Footage Yields a Strong Universal Face Embedder

78% match2025

Wildlife Target Re-Identification Using Self-supervised Learning in Non-Urban Settings

78% match2025

Recurrence over Video Frames (RoVF) for the Re-identification of Meerkats

78% match2024

Cite this paper

@article{miyaguchi2024transfer,
  title  = {Transfer Learning with Self-Supervised Vision Transformers for Snake Identification},
  author = {Anthony Miyaguchi and Murilo Gustineli and Austin Fischer and Ryan Lundqvist},
  year   = {2024},
  eprint = {2407.06178},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2407.06178v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2407.06178.svg)](https://codeofpaper.com/paper/2407.06178)

Discussion