Residual Objectness for Imbalance Reduction

08/24/2019
by   Joya Chen, et al.
0

For a long time, object detectors have suffered from extreme imbalance between foregrounds and backgrounds. While several sampling/reweighting schemes have been explored to alleviate the imbalance, they are usually heuristic and demand laborious hyper-parameters tuning, which is hard to achieve the optimality. In this paper, we first reveal that such the imbalance could be addressed in a learning-based manner. Guided by this illuminating observation, we propose a novel Residual Objectness (ResObj) mechanism that addresses the imbalance by end-to-end optimization, while no further hand-crafted sampling/reweighting is required. Specifically, by applying multiple cascaded objectness-related modules with residual connections, we formulate an elegant consecutive refinement procedure for distinguishing the foregrounds from backgrounds, thereby progressively addressing the imbalance. Extensive experiments present the effectiveness of our method, as well as its compatibility and adaptivity for both region-based and one-stage detectors, namely, the RetinaNet-ResObj, YOLOv3-ResObj and FasterRCNN-ResObj achieve relative 3.6 their vanilla models on COCO, respectively.

READ FULL TEXT
research
09/11/2019

Are Sampling Heuristics Necessary in Object Detectors?

The prevalent object detectors to date, such as Faster R-CNN and RetinaN...
research
06/16/2020

Foreground-Background Imbalance Problem in Deep Object Detectors: A Review

Recent years have witnessed the remarkable developments made by deep lea...
research
11/09/2020

Dual ResGCN for Balanced Scene GraphGeneration

Visual scene graph generation is a challenging task. Previous works have...
research
08/07/2017

Focal Loss for Dense Object Detection

The highest accuracy object detectors to date are based on a two-stage a...
research
12/11/2019

IoU-uniform R-CNN: Breaking Through the Limitations of RPN

Region Proposal Network (RPN) is the cornerstone of two-stage object det...
research
08/25/2023

Dynamic Residual Classifier for Class Incremental Learning

The rehearsal strategy is widely used to alleviate the catastrophic forg...
research
09/21/2019

Generating Positive Bounding Boxes for Balanced Training of Object Detectors

Two-stage deep object detectors generate a set of regions-of-interest (R...

Please sign up or login with your details

Forgot password? Click here to reset