Solving Missing-Annotation Object Detection with Background Recalibration Loss

02/12/2020
by   Han Zhang, et al.
0

This paper focuses on a novel and challenging detection scenario: A majority of true objects/instances is unlabeled in the datasets, so these missing-labeled areas will be regarded as the background during training. Previous art on this problem has proposed to use soft sampling to re-weight the gradients of RoIs based on the overlaps with positive instances, while their method is mainly based on the two-stage detector (i.e. Faster RCNN) which is more robust and friendly for the missing label scenario. In this paper, we introduce a superior solution called Background Recalibration Loss (BRL) that can automatically re-calibrate the loss signals according to the pre-defined IoU threshold and input image. Our design is built on the one-stage detector which is faster and lighter. Inspired by the Focal Loss formulation, we make several significant modifications to fit on the missing-annotation circumstance. We conduct extensive experiments on the curated PASCAL VOC and MS COCO datasets. The results demonstrate that our proposed method outperforms the baseline and other state-of-the-arts by a large margin.

READ FULL TEXT
research
06/18/2018

Soft Sampling for Robust Object Detection

We study the robustness of object detection under the presence of missin...
research
02/11/2020

Object Detection as a Positive-Unlabeled Problem

As with other deep learning methods, label quality is important for lear...
research
07/06/2017

RON: Reverse Connection with Objectness Prior Networks for Object Detection

We present RON, an efficient and effective framework for generic object ...
research
04/21/2022

Absolute Wrong Makes Better: Boosting Weakly Supervised Object Detection via Negative Deterministic Information

Weakly supervised object detection (WSOD) is a challenging task, in whic...
research
03/30/2021

Class-Aware Robust Adversarial Training for Object Detection

Object detection is an important computer vision task with plenty of rea...
research
02/05/2020

Accelerating Object Detection by Erasing Background Activations

Recent advances in deep learning have enabled complex real-world use cas...
research
04/09/2020

Decoupled Gradient Harmonized Detector for Partial Annotation: Application to Signet Ring Cell Detection

Early diagnosis of signet ring cell carcinoma dramatically improves the ...

Please sign up or login with your details

Forgot password? Click here to reset