RefineMask (CVPR 2021)

Related tags

Deep Learninginstance-segmentation
Overview

RefineMask: Towards High-Quality Instance Segmentation
with Fine-Grained Features (CVPR 2021)

This repo is the official implementation of RefineMask: Towards High-Quality Instance Segmentation with Fine-Grained Features.

Framework

image

Main Results

Results on COCO

Method Backbone Schedule AP AP* Checkpoint
Mask R-CNN R50-FPN 1x 34.7 36.8
RefineMask R50-FPN 1x 37.3 40.6 download
Mask R-CNN R50-FPN 2x 35.4 37.7
RefineMask R50-FPN 2x 37.8 41.2 download
Mask R-CNN R101-FPN 1x 36.1 38.4
RefineMask R101-FPN 1x 38.6 41.8 download
Mask R-CNN R101-FPN 2x 36.6 39.3
RefineMask R101-FPN 2x 39.0 42.4 download

Note: No data augmentations except standard horizontal flipping were used.

Results on LVIS

Method Backbone Schedule AP APr APc APf Checkpoint
Mask R-CNN R50-FPN 1x 22.1 10.1 21.7 30.0
RefineMask R50-FPN 1x 25.7 13.8 24.9 31.8 download
Mask R-CNN R101-FPN 1x 23.7 12.3 23.2 29.1
RefineMask R101-FPN 1x 27.1 15.6 26.2 33.1 download

Results on Cityscapes

Method Backbone Schedule AP APS APM APL Checkpoint
Mask R-CNN R50-FPN 1x 33.8 12.0 31.5 51.8
RefineMask R50-FPN 1x 37.6 14.0 35.4 57.9 download

Efficiency of RefineMask

Method AP AP* FPS
Mask R-CNN 34.7 36.8 15.7
PointRend 35.6 38.7 11.4
HTC 37.4 40.7 4.4
RefineMask 37.3 40.9 11.4

Usage

Requirements

  • Python 3.6+
  • Pytorch 1.5.0
  • mmcv-full 1.0.5

Datasets

data
  ├── coco
  |   ├── annotations
  │   │   │   ├── instances_train2017.json
  │   │   │   ├── instances_val2017.json
  │   │   │   ├── lvis_v0.5_val_cocofied.json
  │   ├── train2017
  │   │   ├── 000000004134.png
  │   │   ├── 000000031817.png
  │   │   ├── ......
  │   ├── val2017
  │   ├── test2017
  ├── lvis
  |   ├── annotations
  │   │   │   ├── lvis_v1_train.json
  │   │   │   ├── lvis_v1_val.json
  │   ├── train2017
  │   │   ├── 000000004134.png
  │   │   ├── 000000031817.png
  │   │   ├── ......
  │   ├── val2017
  │   ├── test2017
  ├── cityscapes
  |   ├── annotations
  │   │   │   ├── instancesonly_filtered_gtFine_train.json
  │   │   │   ├── instancesonly_filtered_gtFine_val.json
  │   ├── leftImg8bit
  │   |   ├── train
  │   │   ├── val
  │   │   ├── test

Note: We used the lvis-v1.0 dataset which consists of 1203 categories.

Training

./scripts/dist_train.sh ./configs/refinemask/coco/r50-refinemask-1x.py 8

Note: The codes only support batch size 1 per GPU, and we trained all models with a total batch size 16x1. If you train models with a total batch size 8x1, the performance may drop. We will support batch size 2 or more per GPU later. You can use ./scripts/slurm_train.sh for training with multi-nodes.

Inference

./scripts/dist_test.sh ./configs/refinemask/coco/r50-refinemask-1x.py xxxx.pth 8

Citation

@article{zhang2021refinemask,
  title={RefineMask: Towards High-Quality Instance Segmentation with Fine-Grained Features},
  author={Gang, Zhang and Xin, Lu and Jingru, Tan and Jianmin, Li and Zhaoxiang, Zhang and Quanquan, Li and Xiaolin, Hu},
  journal={arXiv preprint arXiv:2104.08569},
  year={2021}
}
Owner
Gang Zhang
Ph.D. student in Tsinghua University [email protected]
Gang Zhang
GitHub Repository
GULAG: GUessing LAnGuages with neural networks

GULAG: GUessing LAnGuages with neural networks Classify languages in text via neural networks. Привет! My name is Egor. Was für ein herrliches Frühl

Egor Spirin 12 Sep 02, 2022
Tidy interface to polars

tidypolars tidypolars is a data frame library built on top of the blazingly fast polars library that gives access to methods and functions familiar to

Mark Fairbanks 144 Jan 08, 2023
🔎 Super-scale your images and run experiments with Residual Dense and Adversarial Networks.

Image Super-Resolution (ISR) The goal of this project is to upscale and improve the quality of low resolution images. This project contains Keras impl

idealo 4k Jan 08, 2023
The Unsupervised Reinforcement Learning Benchmark (URLB)

The Unsupervised Reinforcement Learning Benchmark (URLB) URLB provides a set of leading algorithms for unsupervised reinforcement learning where agent

259 Dec 26, 2022
Answer a series of contextually-dependent questions like they may occur in natural human-to-human conversations.

SCAI-QReCC-21 [leaderboards] [registration] [forum] [contact] [SCAI] Answer a series of contextually-dependent questions like they may occur in natura

19 Sep 28, 2022
Companion code for the paper "Meta-Learning the Search Distribution of Black-Box Random Search Based Adversarial Attacks" by Yatsura et al.

META-RS This is the companion code for the paper "Meta-Learning the Search Distribution of Black-Box Random Search Based Adversarial Attacks" by Yatsu

Bosch Research 7 Dec 09, 2022
Implementation of the paper "Self-Promoted Prototype Refinement for Few-Shot Class-Incremental Learning"

Self-Promoted Prototype Refinement for Few-Shot Class-Incremental Learning This is the implementation of the paper "Self-Promoted Prototype Refinement

Kai Zhu 78 Dec 02, 2022
AttentionGAN for Unpaired Image-to-Image Translation & Multi-Domain Image-to-Image Translation

AttentionGAN-v2 for Unpaired Image-to-Image Translation AttentionGAN-v2 Framework The proposed generator learns both foreground and background attenti

Hao Tang 530 Dec 27, 2022
Distilling Motion Planner Augmented Policies into Visual Control Policies for Robot Manipulation (CoRL 2021)

Distilling Motion Planner Augmented Policies into Visual Control Policies for Robot Manipulation [Project website] [Paper] This project is a PyTorch i

Cognitive Learning for Vision and Robotics (CLVR) lab @ USC 6 Feb 28, 2022
Explainable Medical ImageSegmentation via GenerativeAdversarial Networks andLayer-wise Relevance Propagation

MedAI: Transparency in Medical Image Segmentation What is this repo This repo contains the code and experiments that are implemented to contribute in

Awadelrahman M. A. Ahmed 1 Nov 22, 2021
Precomputed Real-Time Texture Synthesis with Markovian Generative Adversarial Networks

MGANs Training & Testing code (torch), pre-trained models and supplementary materials for "Precomputed Real-Time Texture Synthesis with Markovian Gene

290 Nov 15, 2022
Our solution for SSN Invente 2021's Hackathon

Our solution for SSN Invente 2021's Hackathon. To help maitain godowns in a pristine and safe condition using raspberry pi.

1 Jan 12, 2022
The code release of paper 'Domain Generalization for Medical Imaging Classification with Linear-Dependency Regularization' NIPS 2020.

Domain Generalization for Medical Imaging Classification with Linear Dependency Regularization The code release of paper 'Domain Generalization for Me

Yufei Wang 56 Dec 28, 2022
CowHerd is a partially-observed reinforcement learning environment

CowHerd is a partially-observed reinforcement learning environment, where the player walks around an area and is rewarded for milking cows. The cows try to escape and the player can place fences to h

Danijar Hafner 6 Mar 06, 2022
A Research-oriented Federated Learning Library and Benchmark Platform for Graph Neural Networks. Accepted to ICLR'2021 - DPML and MLSys'21 - GNNSys workshops.

FedGraphNN: A Federated Learning System and Benchmark for Graph Neural Networks A Research-oriented Federated Learning Library and Benchmark Platform

FedML-AI 175 Dec 01, 2022
A CV toolkit for my papers.

PyTorch-Encoding created by Hang Zhang Documentation Please visit the Docs for detail instructions of installation and usage. Please visit the link to

Hang Zhang 2k Jan 04, 2023
Implementation of "Glancing Transformer for Non-Autoregressive Neural Machine Translation"

GLAT Implementation for the ACL2021 paper "Glancing Transformer for Non-Autoregressive Neural Machine Translation" Requirements Python = 3.7 Pytorch

117 Jan 09, 2023
Torch-mutable-modules - Use in-place and assignment operations on PyTorch module parameters with support for autograd

Torch Mutable Modules Use in-place and assignment operations on PyTorch module p

Kento Nishi 7 Jun 06, 2022
Riemannian Convex Potential Maps

Modeling distributions on Riemannian manifolds is a crucial component in understanding non-Euclidean data that arises, e.g., in physics and geology. The budding approaches in this space are limited b

Facebook Research 61 Nov 28, 2022
Efficiently computes derivatives of numpy code.

Note: Autograd is still being maintained but is no longer actively developed. The main developers (Dougal Maclaurin, David Duvenaud, Matt Johnson, and

Formerly: Harvard Intelligent Probabilistic Systems Group -- Now at Princeton 6.1k Jan 08, 2023