Incremental Object Detection via Meta-Learning

Kj, Joseph and Rajasegaran, Jathushan and Khan, Salman and Khan, Fahad Shahbaz and Balasubramanian, Vineeth N (2021) Incremental Object Detection via Meta-Learning. IEEE Transactions on Pattern Analysis and Machine Intelligence. p. 1. ISSN 0162-8828

[img] Text
IEEE_Transactions_.pdf - Published Version
Available under License Creative Commons Attribution.

Download (7MB)


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. IEEE

[error in script]
IITH Creators:
IITH CreatorsORCiD
Balasubramanian, Vineeth N
Item Type: Article
Uncontrolled Keywords: Deep Neural Networks; Detectors; Feature extraction; Gradient preconditioning; Incremental Learning; Meta-learning; Object detection; Object Detection; Proposals; Standards; Task analysis; Training
Subjects: Computer science
Divisions: Department of Computer Science & Engineering
Depositing User: . LibTrainee 2021
Date Deposited: 29 Aug 2022 11:04
Last Modified: 29 Aug 2022 11:04
Publisher URL:
OA policy:
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 10325 Statistics for this ePrint Item