RODEO: Replay for Online Object Detection

By Manoj Acharya, Tyler L. Hayes, Christopher Kanan. 2020

52 citations recorded by the source corpus.

Humans can incrementally learn to do new visual detection tasks, which is a huge challenge for today's computer vision systems. Incrementally trained deep learning models lack backwards transfer to previously seen classes and suffer from a phenomenon known as $"catastrophic forgetting."$ In this paper, we pioneer online streaming learning for object detection, where an agent must learn examples one at a time with severe memory and computational constraints. In object detection, a system must output all bounding boxes for an image with the correct label. Unlike earlier work, the system described in this paper can learn this task in an online manner with new classes being introduced over time. We achieve this capability by using a novel memory replay mechanism that efficiently replays entire scenes. We achieve state-of-the-art results on both the PASCAL VOC 2007 and MS COCO datasets.

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

Replay in Deep Learning: Current Approaches and Missing Biological Elements (2021, positive): “In Acharya, Hayes, and Kanan (2020) and Liu, Yang, Ravichandran, Bhotika, and Soatto (2020), replay is used to mitigate forgetting for a continual object detection approach.”
Self-Supervised Training Enhances Online Continual Learning (2021, background): “, continual object detection [1, 62], continual semantic segmentation [11, 48], or continual learning for robotics applications [23, 37].”
Online Continual Learning in Image Classification: An Empirical Survey (2021, methodology): “A replay-based method is introduced to address streaming object detection by replaying compressed representation in a fixed memory buffer [131].”
Incremental Object Detection via Meta-Learning (2020, neutral): “[22] proposed to used memory replay to solve a newly formulated online object detection setting, while Joseph et al.”
Selective Replay Enhances Learning in Online Continual Analogical Reasoning (2021, background): “Continual learning has also been studied for object detection [2, 76], semantic segmentation [14, 59], and robotics [24, 50].”
SDDGR: Stable Diffusion-Based Deep Generative Replay for Class Incremental Object Detection (2024, methodology): “Initial researches [1, 11, 35, 60] in class incremental object detection (CIOD) extended image classification methods This CVPR paper is the Open Access version, provided by the Computer Vision Foundation.”
Augmented Box Replay: Overcoming Foreground Shift for Incremental Object Detection (2023, methodology): “While some methods replay images for finetuning [21,27] after training and for meta-learning [28], very few methods replay whole images [52] or stored feature representations [1] during training.”
3rd Continual Learning Workshop Challenge on Egocentric Category and Instance Level Object Understanding (2022, background): “It must be noted that this field is not entirely unexplored from a continual learning perspective [42, 7, 43].”