swMATH ID: 37988
Software Authors: Vincenzo Lomonaco, Davide Maltoni
Description: CORe50: a New Dataset and Benchmark for Continuous Object Recognition. Continuous/Lifelong learning of high-dimensional data streams is a challenging research problem. In fact, fully retraining models each time new data become available is infeasible, due to computational and storage issues, while naïve incremental strategies have been shown to suffer from catastrophic forgetting. In the context of real-world object recognition applications (e.g., robotic vision), where continuous learning is crucial, very few datasets and benchmarks are available to evaluate and compare emerging techniques. In this work we propose a new dataset and benchmark CORe50, specifically designed for continuous object recognition, and introduce baseline approaches for different continuous learning scenarios.
Homepage: https://vlomonaco.github.io/core50/
Source Code:  https://github.com/vlomonaco/core50
Dependencies: Python
Keywords: Computer Vision; Pattern Recognition; arXiv_cs.CV; Artificial Intelligence; arXiv_cs.AI; Machine Learning; arXiv_cs.LG; Robotics; arXiv_cs.RO
Related Software: iCaRL; PyTorch; CIFAR; ImageNet; MNIST; OpenLORIS; Stream-51; Transformers; Caffe; fastai; dtoolAI; Dopamine; OpenAI Gym; TensorFlow; Avalanche; FitNets; Net2Net; Adam; AlexNet; Caltech-UCSD Birds
Cited in: 3 Publications

Citations by Year