change README

xvjiarui · Apr 27, 2019 · 760c9df · 760c9df
1 parent e0d84a6
commit 760c9df
Showing 1 changed file with 56 additions and 46 deletions.
diff --git a/README.md b/README.md
@@ -1,44 +1,74 @@
+# GCNet
 
-# GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond
-By [Yue Cao](http://yue-cao.me), [Jiarui XU](http://jerryxu.net), Stephen Lin, Fangyun Wei, [Han Hu](https://sites.google.com/site/hanhushomepage/).
+By [Yue Cao](http://yue-cao.me), [Jiarui Xu](http://jerryxu.net), [Stephen Lin](https://scholar.google.com/citations?user=c3PYmxUAAAAJ&hl=en), Fangyun Wei, [Han Hu](https://sites.google.com/site/hanhushomepage/).
 
-This repo is an implementation of [GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond](https://arxiv.org/abs/1904.11492) 
-based on open-mmlab's [mmdetection](https://github.com/open-mmlab/mmdetection).
+This repo is a official implementation of ["GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond"](https://arxiv.org/abs/1904.11492) on COCO object detection based on open-mmlab's [mmdetection](https://github.com/open-mmlab/mmdetection).
 Many thanks to mmdetection for their simple and clean framework.
 
-## Abstract
+## Introduction
 
-The Non-Local Network (NLNet) presents a pioneering approach for capturing long-range dependencies, via aggregating query-specific global context to each query position. 
-However, through a rigorous empirical analysis, we have found that the global contexts modeled by non-local network are almost the same for different query positions within an image. 
-In this paper, we take advantage of this finding to create a simplified network based on a query-independent formulation, which maintains the accuracy of NLNet but with significantly less computation. 
-We further observe that this simplified design shares similar structure with Squeeze-Excitation Network (SENet). 
-Hence we unify them into a three-step general framework for global context modeling. 
-Within the general framework, we design a better instantiation, called the global context (GC) block, which is lightweight and can effectively model the global context. 
-The lightweight property allows us to apply it for multiple layers in a backbone network to construct a global context network (GCNet), which generally outperforms both simplified NLNet and SENet on major benchmarks for various recognition tasks.
+GCNet is initially described in [arxiv](https://arxiv.org/abs/1904.11492). 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.
 
-## Visualization
-![vis](figs/vis.png)
-Visualization of attention maps (heatmaps) for different query positions (red points) in a non-local block on COCO object detection. 
-In the same image, the attention maps of different query points are almost the same.
+## Citing GCNet
+
+```
+@article{cao2019GCNet,
+  title={GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond},
+  author={Cao, Yue and Xu, Jiarui and Stephen, Lin and Fangyun, Wei and Hu, Han},
+  journal={arXiv preprint arXiv:1904.11492},
+  year={2019}
+}
+```
+
+## Main Results
+
+| Backbone  |      Model       |    Context     | Lr schd | box AP | mask AP | Download  |
+| :-------: | :--------------: | :------------: | :-----: | :----: | :-----: | :-------: |
+| R-50-FPN  |       Mask       |       -        |   1x    |  37.2  |  33.8   | [model]() |
+|           |                  | GC(c3-c5, r16) |   1x    |  39.4  |  35.7   | [model]() |
+|           |                  | GC(c3-c5, r4)  |   1x    |  39.9  |  36.2   | [model]() |
+| R-101-FPN |       Mask       |       -        |   1x    |  39.8  |  36.0   | [model]() |
+|           |                  | GC(c3-c5, r16) |   1x    |  41.1  |  37.4   | [model]() |
+|           |                  | GC(c3-c5, r4)  |   1x    |  41.7  |  37.6   | [model]() |
+| X-101-FPN |       Mask       |       -        |   1x    |  41.2  |  37.3   | [model]() |
+|           |                  | GC(c3-c5, r16) |   1x    |  42.4  |  38.0   | [model]() |
+|           |                  | GC(c3-c5, r4)  |   1x    |  42.9  |  38.5   | [model]() |
+| X-101-FPN |   Cascade Mask   |       -        |   1x    |  44.7  |  38.3   | [model]() |
+|           |                  | GC(c3-c5, r16) |   1x    |  45.9  |  39.3   | [model]() |
+|           |                  | GC(c3-c5, r4)  |   1x    |  46.5  |  39.7   | [model]() |
+| X-101-FPN | DCN Cascade Mask |       -        |   1x    |  47.1  |  40.4   | [model]() |
+|           |                  | GC(c3-c5, r16) |   1x    |  47.9  |  40.9   | [model]() |
+|           |                  | GC(c3-c5, r4)  |   1x    |  47.9  |  40.8   | [model]() |
+
+**Notes:**
+
+- `GC` denotes Global Context block is inserted after 1x1 conv of backbone. 
+- `DCN` denotes replace 3x3 conv in `c3-c5` stages of backbone.
+- `r4` and `r16` denote ratio 4 and ratio 16 GC block respectively. 
 
 ## Requirements
+
 - Linux(tested on Ubuntu 16.04)
 - Python 3.6+
 - PyTorch 1.0.0
 - Cython
 - [apex](https://github.com/NVIDIA/apex) (Sync BN)
 
 ## Install
+
 a. Install PyTorch 1.0 and torchvision following the [official instructions](https://pytorch.org/).
 
 b. Install latest apex with CUDA and C++ extensions following this [instructions](https://github.com/NVIDIA/apex#quick-start). 
 The [Sync BN](https://nvidia.github.io/apex/parallel.html#apex.parallel.SyncBatchNorm) implemented by apex is required.
 
 c. Clone the GCNet repository. 
+
 ```bash
  git clone https://github.com/xvjiarui/GCNet.git 
- ```
+```
+
 d. Compile cuda extensions.
+
 ```bash
 cd GCNet
 pip install cython  # or "conda install cython" if you prefer conda
@@ -55,7 +85,6 @@ python(3) setup.py install  # add --user if you want to install it locally
 Note: You need to run the last step each time you pull updates from github. 
 Or you can run `python(3) setup.py develop` or `pip install -e .` to install mmdetection if you want to make modifications to it frequently.
 
-
 Please refer to mmdetection install [instruction](https://github.com/open-mmlab/mmdetection/blob/master/INSTALL.md) for more details.
 
 ## Environment
@@ -74,49 +103,30 @@ Please refer to mmdetection install [instruction](https://github.com/open-mmlab/
 - NCCL 2.3.5
 
 ## Usage
+
 ### Train
+
 As in original mmdetection, distributed training is recommended for either single machine or multiple machines.
+
 ```bash
 ./tools/dist_train.sh <CONFIG_FILE> <GPU_NUM> [optional arguments]
 ```
+
 Supported arguments are:
+
 - --validate: perform evaluation every k (default=1) epochs during the training.
 - --work_dir <WORK_DIR>: if specified, the path in config file will be replaced.
 
 ### Evaluation
+
 To evaluate trained models, output file is required.
+
 ```bash
 python tools/test.py <CONFIG_FILE> <MODEL_PATH> [optional arguments]
 ```
+
 Supported arguments are:
+
 - --gpus: number of GPU used for evaluation
 - --out: output file name, usually ends wiht `.pkl`
 - --eval: type of evaluation need, for mask-rcnn, `bbox segm` would evaluate both bounding box and mask AP. 
-
-## Main Results
-
-| Backbone  | Model            | Context        | Lr schd | box AP | mask AP | Download  |
-|:---------:|:----------------:|:--------------:|:-------:|:------:|:-------:|:---------:|
-| R-50-FPN  | Mask             | -              | 1x      | 37.2   | 33.8    | [model]() |
-| R-50-FPN  | Mask             | GC(c3-c5, r16) | 1x      | 39.4   | 35.7    | [model]() |
-| R-50-FPN  | Mask             | GC(c3-c5, r4)  | 1x      | 39.9   | 36.2    | [model]() |
-| R-101-FPN | Mask             | -              | 1x      | 39.8   | 36.0    | [model]() |
-| R-101-FPN | Mask             | GC(c3-c5, r16) | 1x      | 41.1   | 37.4    | [model]() |
-| R-101-FPN | Mask             | GC(c3-c5, r4)  | 1x      | 41.7   | 37.6    | [model]() |
-| X-101-FPN | Mask             | -              | 1x      | 41.2   | 37.3    | [model]() |
-| X-101-FPN | Mask             | GC(c3-c5, r16) | 1x      | 42.4   | 38.0    | [model]() |
-| X-101-FPN | Mask             | GC(c3-c5, r4)  | 1x      | 42.9   | 38.5    | [model]() |
-| R-101-FPN | Cascade Mask     | -              | 1x      | 44.7   | 38.3    | [model]() |
-| R-101-FPN | Cascade Mask     | GC(c3-c5, r16) | 1x      | 45.9   | 39.3    | [model]() |
-| R-101-FPN | Cascade Mask     | GC(c3-c5, r4)  | 1x      | 46.5   | 39.7    | [model]() |
-| R-101-FPN | DCN Cascade Mask | -              | 1x      | 47.1   | 40.4    | [model]() |
-| R-101-FPN | DCN Cascade Mask | GC(c3-c5, r16) | 1x      | 47.9   | 40.9    | [model]() |
-| R-101-FPN | DCN Cascade Mask | GC(c3-c5, r4)  | 1x      | 47.9   | 40.8    | [model]() |
-
-**Notes:**
-- `GC` denotes Global Context block is inserted after 1x1 conv of backbone. 
-- `DCN` denotes replace 3x3 conv in `c3-c5` stages of backbone.
-- `r4` and `r16` denote ratio 4 and ratio 16 GC block respectively. 
-
-## Citation
-