CORe50: a New Dataset and Benchmark for Continuous Object Recognition

2017/05/09 by Vincenzo Lomonaco, Lomonaco, Vincenzo, Davide Maltoni +1 · 17 citations
Computer Science · #Domain Adaptation and Few-Shot Learning #Machine Learning and ELM #Advanced Neural Network Applications

paper · pdf · doi:10.48550/arxiv.1705.03550

Abstract

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.

Cited by

• Continual Learning at the Edge: An Agnostic IIoT Architecture
• Open Horizons: Evaluating Deep Models in the Wild
• Reconstructing Objects along Hand Interaction Timelines in Egocentric Video
• Minimizing Hyperbolic Embedding Distortion with LLM-Guided Hierarchy Restructuring
• Continual Knowledge Consolidation LORA for Domain Incremental Learning
• CaMiT: A Time-Aware Car Model Dataset for Classification and Generation
• Domain Generalizable Continual Learning
• Your VAR Model is Secretly an Efficient and Explainable Generative Classifier
• Personalized Vision via Visual In-Context Learning
• Simulated Cortical Magnification Supports Self-Supervised Object Learning
• MOCHA: Multi-modal Objects-aware Cross-arcHitecture Alignment
• CIFNet: An Analytic Neural Learning Framework for Efficient and Calibrated Class-Incremental Learning
• RICO: Two Realistic Benchmarks and an In-Depth Analysis for Incremental Learning in Object Detection
• SAGE: Scale-Aware Gradual Evolution for Continual Knowledge Graph Embedding
• Towards Efficient Prompt-based Continual Learning in Distributed Medical AI
• Statistical Theory of Multi-stage Newton Iteration Algorithm for Online Continual Learning
• Continual Learning for VLMs: A Survey and Taxonomy Beyond Forgetting

Related

• OpenLORIS-Object: A Robotic Vision Dataset and Benchmark for Lifelong Deep Learning
• 3rd Continual Learning Workshop Challenge on Egocentric Category and Instance Level Object Understanding
• Rehearsal-Free Continual Learning over Small Non-I.I.D. Batches
• CVPR 2020 Continual Learning in Computer Vision Competition: Approaches,\n Results, Current Challenges and Future Directions
• Continual-Learning-as-a-Service (CLaaS): On-Demand Efficient Adaptation of Predictive Models