Robust Object Detection via Instance-Level Temporal Cycle Confusion

Xin Wang, Thomas E. Huang, Benlin Liu, Fisher Yu, Xiaolong Wang, Joseph E. Gonzalez, Trevor Darrell
ICCV 2021

Robust Object Detection via Instance-Level Temporal Cycle Confusion

Abstract

Building reliable object detectors that are robust to domain shifts, such as various changes in context, viewpoint, and object appearances, is critical for real-world applications. In this work, we study the effectiveness of auxiliary self-supervised tasks to improve the out-of-distribution generalization of object detectors. Inspired by the principle of maximum entropy, we introduce a novel self-supervised task, instance-level temporal cycle confusion (CycConf), which operates on the region features of the object detectors. For each object, the task is to find the most different object proposals in the adjacent frame in a video and then cycle back to itself for self-supervision. CycConf encourages the object detector to explore invariant structures across instances under various motions, which leads to improved model robustness in unseen domains at test time. We observe consistent out-of-domain performance improvements when training object detectors in tandem with self-supervised tasks on large-scale video datasets (BDD100K and Waymo open data). The joint training framework also establishes a new state-of-the-art on standard unsupervised domain adaptative detection benchmarks (Cityscapes, Foggy Cityscapes, and Sim10K).

Video

Paper

Code

paper
github.com/xinw1012/cycle-confusion

Citation

@inproceedings{wang2021robust,
    title     = {Robust Object Detection via Instance-Level Temporal Cycle Confusion}, 
    author    = {Xin Wang and Thomas E. Huang and Benlin Liu and Fisher Yu and Xiaolong Wang and Joseph E. Gonzalez and Trevor Darrell},
    booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
    year      = {2021}
}

Related


Warp Consistency for Unsupervised Learning of Dense Correspondences

Warp Consistency for Unsupervised Learning of Dense Correspondences

ICCV 2021 Oral We propose Warp Consistency, an unsupervised learning objective for dense correspondence regression.


Frustratingly Simple Few-Shot Object Detection

Frustratingly Simple Few-Shot Object Detection

ICML 2020 State-of-the-art few-shot detection method with backpropagation learning.


Few Shot Object Detection via Feature Reweighting

Few Shot Object Detection via Feature Reweighting

ICCV 2019 We develop a few-shot object detector that can learn to detect novel objects from only a few annotated examples.


Characterizing Adversarial Examples Based on Spatial Consistency Information for Semantic Segmentation

Characterizing Adversarial Examples Based on Spatial Consistency Information for Semantic Segmentation

ECCV 2018 We aim to characterize adversarial examples based on spatial context information in semantic segmentation.