F-SIOL-310: A Robotic Dataset and Benchmark for Few-Shot Incremental Object Learning

By Ali Ayub, Alan R. Wagner. 2021

14 citations recorded by the source corpus.

Deep learning has achieved remarkable success in object recognition tasks through the availability of large scale datasets like ImageNet. However, deep learning systems suffer from catastrophic forgetting when learning incrementally without replaying old data. For real-world applications, robots also need to incrementally learn new objects. Further, since robots have limited human assistance available, they must learn from only a few examples. However, very few object recognition datasets and benchmarks exist to test incremental learning capability for robotic vision. Further, there is no dataset or benchmark specifically designed for incremental object learning from a few examples. To fill this gap, we present a new dataset termed F-SIOL-310 (Few-Shot Incremental Object Learning) which is specifically captured for testing few-shot incremental object learning capability for robotic vision. We also provide benchmarks and evaluations of 8 incremental learning algorithms on F-SIOL-310 for future comparisons. Our results demonstrate that the few-shot incremental object learning problem for robotic vision is far from being solved.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

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

In Artificial Intelligence

1. Tyler L. Hayes
2. Christopher Kanan
3. Ali Ayub
4. C. Fendley
5. Alan R. Wagner
6. Xin Ning
7. Weijun Li
8. Prayag Tiwari

In Computer Vision and Pattern Recognition

1. Tyler L. Hayes
2. Christopher Kanan
3. C. Fendley
4. Xin Ning
5. Weijun Li
6. Ali Ayub
7. Prayag Tiwari
8. Hang Ran

Citations in the corpus

• Few-Shot Continual Active Learning by a Robot (2022, background): “A few works have been proposed for FSCIL, to continually learn from a few-examples per class using limited memory [10, 9, 32, 33].”
• Online Continual Learning for Embedded Devices (2022, methodology): “To perform online continual learning experiments using this dataset, we use the class-iid data ordering in both 5-Shot and 10-Shot learning scenarios, as suggested by Ayub & Wagner (2021).”
• LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning (2023, unknown): “F-SIOL-310 [3] and OpenLORIS [61] are challenging real-world lifelong object learning datasets that are captured from robotic vision systems.”
• A Survey on Few-Shot Class-Incremental Learning (2023, background): “However, the few-shot incremental object learning problem for robotic vision remains unresolved (Ayub and Wagner, 2021).”