Multi-Task Incremental Learning for Object Detection

Abstract

Multi-task learns multiple tasks, while sharing knowledge and computation among them. However, it suffers from catastrophic forgetting of previous knowledge when learned incrementally without access to the old data. Most existing object detectors are domain-specific and static, while some are learned incrementally but only within a single domain. Training an object detector incrementally across various domains has rarely been explored. In this work, we propose three incremental learning scenarios across various domains and categories for object detection. To mitigate catastrophic forgetting, attentive feature distillation is proposed to leverages both bottom-up and top-down attentions to extract important information for distillation. We then systematically analyze the proposed distillation method in different scenarios. We find out that, contrary to common understanding, domain gaps have smaller negative impact on incremental detection, while category differences are problematic. For the difficult cases, where the domain gaps and especially category differences are large, we explore three different exemplar sampling methods and show the proposed adaptive sampling method is effective to select diverse and informative samples from entire datasets, to further prevent forgetting. Experimental results show that we achieve the significant improvement in three different scenarios across seven object detection benchmark datasets.

References in corpus (16)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Distilling the Knowledge in a Neural Network
• Overcoming catastrophic forgetting in neural networks
• YOLOv3: An Incremental Improvement
• DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition
• FitNets: Hints for Thin Deep Nets
• A continual learning survey: Defying forgetting in classification tasks
• Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfer
• Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigence
• Continual Learning Through Synaptic Intelligence
• Incremental Object Detection via Meta-Learning
• Life-Long Disentangled Representation Learning with Cross-Domain Latent Homologies
• Memory Replay GANs: learning to generate images from new categories without forgetting
• RODEO: Replay for Online Object Detection
• Lifelong Object Detection
• Two-Level Residual Distillation based Triple Network for Incremental Object Detection

Cited by in corpus (6)

• Uncertainty-aware Contrastive Distillation for Incremental Semantic Segmentation
• Comprehensive Attention Self-Distillation for Weakly-Supervised Object Detection
• Inherit with Distillation and Evolve with Contrast: Exploring Class Incremental Semantic Segmentation Without Exemplar Memory
• Replay in Deep Learning: Current Approaches and Missing Biological Elements
• Lifelong Learning using Eigentasks: Task Separation, Skill Acquisition, and Selective Transfer
• Online Continual Learning in Image Classification: An Empirical Survey