DeepAI AI Chat
Log In Sign Up

Location-Sensitive Visual Recognition with Cross-IOU Loss

04/11/2021
by   Kaiwen Duan, et al.
HUAWEI Technologies Co., Ltd.
University of the Chinese Academy of Sciences
Microsoft
0

Object detection, instance segmentation, and pose estimation are popular visual recognition tasks which require localizing the object by internal or boundary landmarks. This paper summarizes these tasks as location-sensitive visual recognition and proposes a unified solution named location-sensitive network (LSNet). Based on a deep neural network as the backbone, LSNet predicts an anchor point and a set of landmarks which together define the shape of the target object. The key to optimizing the LSNet lies in the ability of fitting various scales, for which we design a novel loss function named cross-IOU loss that computes the cross-IOU of each anchor point-landmark pair to approximate the global IOU between the prediction and ground-truth. The flexibly located and accurately predicted landmarks also enable LSNet to incorporate richer contextual information for visual recognition. Evaluated on the MS-COCO dataset, LSNet set the new state-of-the-art accuracy for anchor-free object detection (a 53.5 shows promising performance in detecting multi-scale human poses. Code is available at https://github.com/Duankaiwen/LSNet

READ FULL TEXT

page 1

page 8

page 10

04/14/2021

HoughNet: Integrating near and long-range evidence for visual detection

This paper presents HoughNet, a one-stage, anchor-free, voting-based, bo...
11/11/2018

Improved Visual Relocalization by Discovering Anchor Points

We address the visual relocalization problem of predicting the location ...
07/27/2020

Corner Proposal Network for Anchor-free, Two-stage Object Detection

The goal of object detection is to determine the class and location of o...
03/20/2020

CentripetalNet: Pursuing High-quality Keypoint Pairs for Object Detection

Keypoint-based detectors have achieved pretty-well performance. However,...
11/16/2018

DeRPN: Taking a further step toward more general object detection

Most current detection methods have adopted anchor boxes as regression r...
11/01/2021

Feature Aggregation and Refinement Network for 2D AnatomicalLandmark Detection

Localization of anatomical landmarks is essential for clinical diagnosis...
05/23/2021

VS-Net: Voting with Segmentation for Visual Localization

Visual localization is of great importance in robotics and computer visi...

Code Repositories

LSNet

Location-Sensitive Visual Recognition with Cross-IOU Loss


view repo