Incremental Object Detection via Meta-Learning
IEEE Transactions on Pattern Analysis and Machine Intelligence
In a real-world setting, object instances from new classes can be continuously encountered by object detectors. When existing object detectors are applied to such scenarios, their performance on old classes deteriorates significantly. A few efforts have been reported to address this limitation, all of which apply variants of knowledge distillation to avoid catastrophic forgetting. We note that although distillation helps to retain previous learning, it obstructs fast adaptability to new tasks, which is a critical requirement for incremental learning. In this pursuit, we propose a meta-learning approach that learns to reshape model gradients, such that information across incremental tasks is optimally shared. This ensures a seamless information transfer via a meta-learned gradient preconditioning that minimizes forgetting and maximizes knowledge transfer. In comparison to existing meta-learning methods, our approach is task-agnostic, allows incremental addition of new-classes and scales to high-capacity models for object detection. We evaluate our approach on a variety of incremental learning settings defined on PASCAL-VOC and MS COCO datasets, where our approach performs favourably well against state-of-the-art methods. Code and trained models: https://github.com/JosephKJ/iOD. © 2020, CC BY.
Deep neural networks, Distillation, Knowledge management, Object recognition, Transfer learning, Catastrophic forgetting, Gradient preconditioning, Incremental learning, Learn+, Meta-learning approach, Metalearning, Object detectors, Objects detection, Performance, Real world setting, Object detection, Computer Vision and Pattern Recognition (cs.CV), Image and Video Processing (eess.IV), Machine Learning (cs.LG), Machine Learning (stat.ML)
J. Kj, J. Rajasegaran, S. Khan, F. S. Khan and V. N Balasubramanian, "Incremental Object Detection via Meta-Learning," in IEEE Transactions on Pattern Analysis and Machine Intelligence, doi: 10.1109/TPAMI.2021.3124133.