Skip to content
Branch: master
Find file History
hellock Make data pre-processing pipeline customizable (#935)
* define data pipelines

* update two config files

* minor fix for config files

* allow img_scale to be optional and update config

* add some docstrings

* add extra aug to transform

* bug fix for mask resizing

* fix cropping

* add faster rcnn example

* fix imports

* fix robustness testing

* add img_norm_cfg to img_meta

* fix the inference api with the new data pipeline

* fix proposal loading

* delete args of DefaultFormatBundle

* add more configs

* update configs

* bug fix

* add a brief doc

* update gt_labels in RandomCrop

* fix key error for new apis

* bug fix for masks of crowd bboxes

* add argument data_root

* minor fix

* update new hrnet configs

* update docs

* rename MultiscaleFlipAug to MultiScaleFlipAug

* add __repr__ for all transforms

* move DATA_PIPELINE.md to docs/

* fix image url
Latest commit 0d5233a Aug 23, 2019

README.md

GCNet for Object Detection

By Yue Cao, Jiarui Xu, Stephen Lin, Fangyun Wei, Han Hu.

We provide config files to reproduce the results in the paper for "GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond" on COCO object detection.

Introduction

GCNet is initially described in arxiv. Via absorbing advantages of Non-Local Networks (NLNet) and Squeeze-Excitation Networks (SENet), GCNet provides a simple, fast and effective approach for global context modeling, which generally outperforms both NLNet and SENet on major benchmarks for various recognition tasks.

Citing GCNet

@article{cao2019GCNet,
  title={GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond},
  author={Cao, Yue and Xu, Jiarui and Lin, Stephen and Wei, Fangyun and Hu, Han},
  journal={arXiv preprint arXiv:1904.11492},
  year={2019}
}

Results and models

The results on COCO 2017val are shown in the below table.

Backbone Model Context Lr schd Mem (GB) Train time (s/iter) Inf time (fps) box AP mask AP Download
R-50-FPN Mask GC(c3-c5, r16) 1x 4.5 0.533 10.1 38.5 35.1 model
R-50-FPN Mask GC(c3-c5, r4) 1x 4.6 0.533 9.9 38.9 35.5 model
R-101-FPN Mask GC(c3-c5, r16) 1x 7.0 0.731 8.6 40.8 37.0 model
R-101-FPN Mask GC(c3-c5, r4) 1x 7.1 0.747 8.6 40.8 36.9 model
Backbone Model Context Lr schd Mem (GB) Train time (s/iter) Inf time (fps) box AP mask AP Download
R-50-FPN Mask - 1x 3.9 0.543 10.2 37.2 33.8 model
R-50-FPN Mask GC(c3-c5, r16) 1x 4.5 0.547 9.9 39.4 35.7 model
R-50-FPN Mask GC(c3-c5, r4) 1x 4.6 0.603 9.4 39.9 36.2 model
R-101-FPN Mask - 1x 5.8 0.665 9.2 39.8 36.0 model
R-101-FPN Mask GC(c3-c5, r16) 1x 7.0 0.778 9.0 41.1 37.4 model
R-101-FPN Mask GC(c3-c5, r4) 1x 7.1 0.786 8.9 41.7 37.6 model
X-101-FPN Mask - 1x 7.1 0.912 8.5 41.2 37.3 model
X-101-FPN Mask GC(c3-c5, r16) 1x 8.2 1.055 7.7 42.4 38.0 model
X-101-FPN Mask GC(c3-c5, r4) 1x 8.3 1.037 7.6 42.9 38.5 model
X-101-FPN Cascade Mask - 1x - - - 44.7 38.3 model
X-101-FPN Cascade Mask GC(c3-c5, r16) 1x - - - 45.9 39.3 model
X-101-FPN Cascade Mask GC(c3-c5, r4) 1x - - - 46.5 39.7 model
X-101-FPN DCN Cascade Mask - 1x - - - 47.1 40.4 model
X-101-FPN DCN Cascade Mask GC(c3-c5, r16) 1x - - - 47.9 40.9 model
X-101-FPN DCN Cascade Mask GC(c3-c5, r4) 1x - - - 47.9 40.8 model

Notes:

  • The SyncBN is added in the backbone for all models in Table 2.
  • GC denotes Global Context (GC) block is inserted after 1x1 conv of backbone.
  • DCN denotes replace 3x3 conv with 3x3 Deformable Convolution in c3-c5 stages of backbone.
  • r4 and r16 denote ratio 4 and ratio 16 in GC block respectively.
You can’t perform that action at this time.