CVPR 2020 Continual Learning in Computer Vision Competition: Approaches, Results, Current Challenges and Future Directions

By Vincenzo Lomonaco, Lorenzo Pellegrini, Pau Rodr'iguez L'opez, Massimo Caccia, Qi She, Yu Chen, Quentin Jodelet, Ruiping Wang, Zheda Mai, David Vázquez and 5 more. 2020

54 citations recorded by the source corpus.

In the last few years, we have witnessed a renewed and fast-growing interest in continual learning with deep neural networks with the shared objective of making current AI systems more adaptive, efficient and autonomous. However, despite the significant and undoubted progress of the field in addressing the issue of catastrophic forgetting, benchmarking different continual learning approaches is a difficult task by itself. In fact, given the proliferation of different settings, training and evaluation protocols, metrics and nomenclature, it is often tricky to properly characterize a continual learning algorithm, relate it to other solutions and gauge its real-world applicability. The first Continual Learning in Computer Vision challenge held at CVPR in 2020 has been one of the first opportunities to evaluate different continual learning algorithms on a common hardware with a large set of shared evaluation metrics and 3 different settings based on the realistic CORe50 video benchmark. In this paper, we report the main results of the competition, which counted more than 79 teams registered, 11 finalists and 2300$ in prizes. We also summarize the winning approaches, current challenges and future research directions.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Meta & Continual Learning

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

In Computer Vision and Pattern Recognition

In Meta & Continual Learning

Citations in the corpus

Simpler is Better: off-the-shelf Continual Learning Through Pretrained Backbones (2022, background): “This can be confirmed by the CVPR 2020 Continual Learning Challenge summary report [14], where the authors noted that all the methods proposed solutions leveraging pretrained backbones.”
Continual Learning via Local Module Composition (2021, background): “The rapid growth of continual learning has lead researchers to work on empirical studies [20, 58, 56], surveys [32, 42, 55, 66, 67] as well as CL-specific software [72, 22, 59].”
Recent Advances of Continual Learning in Computer Vision: An Overview (2021, neutral): “Additionally, several competitions [9, 10] related to continual learning in computer vision have been held in both 2020 and 2021.”
A TinyML Platform for On-Device Continual Learning With Quantized Latent Replays (2021, methodology): “In the recent CL challenge at CVPR2020 on the Core50 image dataset, ∼90% of the competitors used rehearsal strategies [16].”
Practical Recommendations for Replay-based Continual Learning Methods (2022, background): “Among them, Replay approaches have empirically proved to be the most effective ones [16].”
TiC-CLIP: Continual Training of CLIP Models (2023, unknown): “Simple replay-based baselines can surpass various methods on standard benchmarks (Lomonaco et al., 2022; Balaji et al., 2020; Prabhu et al., 2020).”
3rd Continual Learning Workshop Challenge on Egocentric Category and Instance Level Object Understanding (2022, methodology): “The competition was organized using the two-phase structure following previous editions of the CLVision workshop [16]: in the pre-selection phase, which took place between the 30th of March 2022 and the 29th of May 2022, participants were asked to submit their solution as a copy of the prediction ob”