DeepAI AI Chat
Log In Sign Up

Identification of Novel Classes for Improving Few-Shot Object Detection

by   Zeyu Shangguan, et al.
University of Southern California

Conventional training of deep neural networks requires a large number of the annotated image which is a laborious and time-consuming task, particularly for rare objects. Few-shot object detection (FSOD) methods offer a remedy by realizing robust object detection using only a few training samples per class. An unexplored challenge for FSOD is that instances from unlabeled novel classes that do not belong to the fixed set of training classes appear in the background. These objects behave similarly to label noise, leading to FSOD performance degradation. We develop a semi-supervised algorithm to detect and then utilize these unlabeled novel objects as positive samples during training to improve FSOD performance. Specifically, we propose a hierarchical ternary classification region proposal network (HTRPN) to localize the potential unlabeled novel objects and assign them new objectness labels. Our improved hierarchical sampling strategy for the region proposal network (RPN) also boosts the perception ability of the object detection model for large objects. Our experimental results indicate that our method is effective and outperforms the existing state-of-the-art (SOTA) FSOD methods.


page 3

page 4

page 5


Improved Region Proposal Network for Enhanced Few-Shot Object Detection

Despite significant success of deep learning in object detection tasks, ...

Few-shot Object Detection

In this paper, we study object detection using a large pool of unlabeled...

PatchContrast: Self-Supervised Pre-training for 3D Object Detection

Accurately detecting objects in the environment is a key challenge for a...

Factors in Finetuning Deep Model for object detection

Finetuning from a pretrained deep model is found to yield state-of-the-a...

Prototypical Region Proposal Networks for Few-Shot Localization and Classification

Recently proposed few-shot image classification methods have generally f...

Co-occurrence matrix analysis-based semi-supervised training for object detection

One of the most important factors in training object recognition network...

Self-Trained Proposal Networks for the Open World

Deep learning-based object proposal methods have enabled significant adv...