Code release for Hu et al., Learning to Segment Every Thing. in CVPR, 2018.

Overview

Learning to Segment Every Thing

This repository contains the code for the following paper:

  • R. Hu, P. Dollár, K. He, T. Darrell, R. Girshick, Learning to Segment Every Thing. in CVPR, 2018. (PDF)
@inproceedings{hu2018learning,
  title={Learning to Segment Every Thing},
  author={Hu, Ronghang and Dollár, Piotr and He, Kaiming and Darrell, Trevor and Girshick, Ross},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  year={2018}
}

Project Page: http://ronghanghu.com/seg_every_thing

Note: this repository is built upon the Detectron codebase for object detection and segmentation (https://github.com/facebookresearch/Detectron), based on Detectron commit 3c4c7f67d37eeb4ab15a87034003980a1d259c94. Please see README_DETECTRON.md for details.

Installation

The installation procedure follows Detectron.

Please find installation instructions for Caffe2 and Detectron in INSTALL.md.

Note: all the experiments below run on 8 GPUs on a single machine. If you have less than 8 GPU available, please modify the yaml config files according to the linear scaling rule. For example, if you only have 4 GPUs, set NUM_GPUS to 4, downscale SOLVER.BASE_LR by 0.5x and multiply SOLVER.STEPS and SOLVER.MAX_ITER by 2x.

Part 1: Controlled Experiments on the COCO dataset

In this work, we explore our approach in two settings. First, we use the COCO dataset to simulate the partially supervised instance segmentation task as a means of establishing quantitative results on a dataset with high-quality annotations and evaluation metrics. Specifically, we split the full set of COCO categories into a subset with mask annotations and a complementary subset for which the system has access to only bounding box annotations. Because the COCO dataset involves only a small number (80) of semantically well-separated classes, quantitative evaluation is precise and reliable.

In our experiments, we split COCO into either

  • VOC Split: 20 PASCAL-VOC classes v.s. 60 non-PASCAL-VOC classes. We experiment with 1) VOC -> non-VOC, where set A={VOC} and 2) non-VOC -> VOC, where set A={non-VOC}.
  • Random Splits: randomly partitioned two subsets A and B of the 80 COCO classes.

and experiment with two training setups:

  • Stage-wise training, where first a Faster R-CNN detector is trained and kept frozen, and then the mask branch (including the weight transfer function) is added later.
  • End-to-end training, where the RPN, the box head, the mask head and the weight transfer function are trained together.

Please refer to Section 4 of our paper for details on the COCO experiments.

COCO Installation: To run the COCO experiments, first download the COCO dataset and install it according to the dataset guide.

Evaluation

The following experiments correspond to the results in Section 4.2 and Table 2 of our paper.

To run the experiments:

  1. Split the COCO dataset into VOC / non-VOC classes:
    python2 lib/datasets/bbox2mask_dataset_processing/coco/split_coco_dataset_voc_nonvoc.py.
  2. Set the training split using SPLIT variable:
  • To train on VOC -> non-VOC, where set A={VOC}, use export SPLIT=voc2nonvoc.
  • To train on non-VOC -> VOC, where set A={non-VOC}, use export SPLIT=nonvoc2voc.

Then use tools/train_net.py to run the following yaml config files for each experiment with ResNet-50-FPN backbone or ResNet-101-FPN backbone.

Please follow the instruction in GETTING_STARTED.md to train with the config files. The training scripts automatically test the trained models and print the bbox and mask APs on the VOC ('coco_split_voc_2014_minival') and non-VOC splits ('coco_split_nonvoc_2014_minival').

Using ResNet-50-FPN backbone:

  1. Class-agnostic (baseline): configs/bbox2mask_coco/${SPLIT}/eval_e2e/e2e_baseline.yaml
  2. MaskX R-CNN (ours, tansfer+MLP): configs/bbox2mask_coco/${SPLIT}/eval_e2e/e2e_clsbox_2_layer_mlp_nograd.yaml
  3. Fully-supervised (oracle): configs/bbox2mask_coco/oracle/e2e_mask_rcnn_R-50-FPN_1x.yaml

Using ResNet-101-FPN backbone:

  1. Class-agnostic (baseline): configs/bbox2mask_coco/${SPLIT}/eval_e2e_R101/e2e_baseline.yaml
  2. MaskX R-CNN (ours, tansfer+MLP): configs/bbox2mask_coco/${SPLIT}/eval_e2e_R101/e2e_clsbox_2_layer_mlp_nograd.yaml
  3. Fully-supervised (oracle): configs/bbox2mask_coco/oracle/e2e_mask_rcnn_R-101-FPN_1x.yaml

Ablation Study

This section runs ablation studies on the VOC Split (20 PASCAL-VOC classes v.s. 60 non-PASCAL-VOC classes) using ResNet-50-FPN backbone. The results correspond to Section 4.1 and Table 1 of our paper.

To run the experiments:

  1. (If you haven't done so in the above section) Split the COCO dataset into VOC / non-VOC classes:
    python2 lib/datasets/bbox2mask_dataset_processing/coco/split_coco_dataset_voc_nonvoc.py.
  2. For Study 1, 2, 3 and 5, download the pre-trained Faster R-CNN model with ResNet-50-FPN by running
    bash lib/datasets/data/trained_models/fetch_coco_faster_rcnn_model.sh.
    (Alternatively, you can train it yourself using configs/12_2017_baselines/e2e_faster_rcnn_R-50-FPN_1x.yaml and copy it to lib/datasets/data/trained_models/28594643_model_final.pkl.)
  3. For Study 1, add the GloVe and random embeddings of the COCO class names to the Faster R-CNN weights with
    python2 lib/datasets/bbox2mask_dataset_processing/coco/add_embeddings_to_weights.py.
  4. Set the training split using SPLIT variable:
  • To train on VOC -> non-VOC, where set A={VOC}, use export SPLIT=voc2nonvoc.
  • To train on non-VOC -> VOC, where set A={non-VOC}, use export SPLIT=nonvoc2voc.

Then use tools/train_net.py to run the following yaml config files for each experiment.

Study 1: Ablation on the input to the weight transfer function (Table 1a)

  • transfer w/ randn: configs/bbox2mask_coco/${SPLIT}/ablation_input/randn_2_layer.yaml
  • transfer w/ GloVe: configs/bbox2mask_coco/${SPLIT}/ablation_input/glove_2_layer.yaml
  • transfer w/ cls: configs/bbox2mask_coco/${SPLIT}/ablation_input/cls_2_layer.yaml
  • transfer w/ box: configs/bbox2mask_coco/${SPLIT}/ablation_input/box_2_layer.yaml
  • transfer w/ cls+box: configs/bbox2mask_coco/${SPLIT}/eval_sw/clsbox_2_layer.yaml
  • class-agnostic (baseline): configs/bbox2mask_coco/${SPLIT}/eval_sw/baseline.yaml
  • fully supervised (oracle): configs/bbox2mask_coco/oracle/mask_rcnn_frozen_features_R-50-FPN_1x.yaml

Study 2: Ablation on the structure of the weight transfer function (Table 1b)

  • transfer w/ 1-layer, none: configs/bbox2mask_coco/${SPLIT}/ablation_structure/clsbox_1_layer.yaml
  • transfer w/ 2-layer, ReLU: configs/bbox2mask_coco/${SPLIT}/ablation_structure/relu/clsbox_2_layer_relu.yaml
  • transfer w/ 2-layer, LeakyReLU: same as 'transfer w/ cls+box' in Study 1
  • transfer w/ 3-layer, ReLU: configs/bbox2mask_coco/${SPLIT}/ablation_structure/relu/clsbox_3_layer_relu.yaml
  • transfer w/ 3-layer, LeakyReLU: configs/bbox2mask_coco/${SPLIT}/ablation_structure/clsbox_3_layer.yaml

Study 3: Impact of the MLP mask branch (Table 1c)

  • class-agnostic: same as 'class-agnostic (baseline)' in Study 1
  • class-agnostic+MLP: configs/bbox2mask_coco/${SPLIT}/ablation_mlp/baseline_mlp.yaml
  • transfer: same as 'transfer w/ cls+box' in Study 1
  • transfer+MLP: configs/bbox2mask_coco/${SPLIT}/ablation_mlp/clsbox_2_layer_mlp.yaml

Study 4: Ablation on the training strategy (Table 1d)

  • class-agnostic + sw: same as 'class-agnostic (baseline)' in Study 1
  • transfer + sw: same as 'transfer w/ cls+box' in Study 1
  • class-agnostic + e2e: configs/bbox2mask_coco/${SPLIT}/eval_e2e/e2e_baseline.yaml
  • transfer + e2e: configs/bbox2mask_coco/${SPLIT}/ablation_e2e_stopgrad/e2e_clsbox_2_layer.yaml
  • transfer + e2e + stopgrad: configs/bbox2mask_coco/${SPLIT}/ablation_e2e_stopgrad/e2e_clsbox_2_layer_nograd.yaml

Study 5: Comparison of random A/B splits (Figure 3)

Note: this ablation study takes a HUGE amount of computation power. It consists of 50 training experiments (= 5 trials * 5 class-number in set A (20/30/40/50/60) * 2 settings (ours/baseline) ), and each training experiment takes approximately 9 hours to complete on 8 GPUs.

Before running Study 5:

  1. Split the COCO dataset into random class splits (This should take a while):
    python2 lib/datasets/bbox2mask_dataset_processing/coco/split_coco_dataset_randsplits.py.
  2. Set the training split using SPLIT variable (e.g. export SPLIT=E1_A20B60). The split has the format E%d_A%dB%d for example, E1_A20B60 is trial No. 1 with 20 random classes in set A and 60 random classes in set B. There are 5 trials (E1 to E5), with 20/30/40/50/60 random classes in set A (A20B60 to A60B20), yielding altogether 25 splits from E1_A20B60 to E5_A60B20.

Then use tools/train_net.py to run the following yaml config files for each experiment.

  • class-agnostic (baseline): configs/bbox2mask_coco/randsplits/eval_sw/${SPLIT}_baseline.yaml
  • tansfer w/ cls+box, 2-layer, LeakyReLU: configs/bbox2mask_coco/randsplits/eval_sw/${SPLIT}_clsbox_2_layer.yaml

Part 2: Large-scale Instance Segmentation on the Visual Genome dataset

In our second setting, we train a large-scale instance segmentation model on 3000 categories using the Visual Genome (VG) dataset. On the Visual Genome dataset, set A (w/ mask data) is the 80 COCO classes, while set B (w/o mask data, only bbox) is the remaining Visual Genome classes that are not in COCO.

Please refer to Section 5 of our paper for details on the Visual Genome experiments.

Inference

To run inference, download the pre-trained final model weights by running:
bash lib/datasets/data/trained_models/fetch_vg3k_final_model.sh
(Alternatively, you may train these weights yourself following the training section below.)

Then, use tools/infer_simple.py for prediction. Note: due to the large number of classes and the model loading overhead, prediction on the first image can take a while.

Using ResNet-50-FPN backbone:

python2 tools/infer_simple.py \
    --cfg configs/bbox2mask_vg/eval_sw/runtest_clsbox_2_layer_mlp_nograd.yaml \
    --output-dir /tmp/detectron-visualizations-vg3k \
    --image-ext jpg \
    --thresh 0.5 --use-vg3k \
    --wts lib/datasets/data/trained_models/33241332_model_final_coco2vg3k_seg.pkl \
    demo_vg3k

Using ResNet-101-FPN backbone:

python2 tools/infer_simple.py \
    --cfg configs/bbox2mask_vg/eval_sw_R101/runtest_clsbox_2_layer_mlp_nograd_R101.yaml \
    --output-dir /tmp/detectron-visualizations-vg3k-R101 \
    --image-ext jpg \
    --thresh 0.5 --use-vg3k \
    --wts lib/datasets/data/trained_models/33219850_model_final_coco2vg3k_seg.pkl \
    demo_vg3k

Training

Visual Genome Installation: To run the Visual Genome experiments, first download the Visual Genome dataset and install it according to the dataset guide. Then download the converted Visual Genome json dataset files (in COCO-format) by running:
bash lib/datasets/data/vg3k_bbox2mask/fetch_vg3k_json.sh.
(Alternatively, you may build the COCO-format json dataset files yourself using the scripts in lib/datasets/bbox2mask_dataset_processing/vg/)

Here, we adopt the stage-wise training strategy as mentioned in Section 5 of our paper. First in Stage 1, a Faster R-CNN detector is trained on all the 3k Visual Genome classes (set A+B). Then in Stage 2, the mask branch (with the weight transfer function) is added and trained on the mask data of the 80 COCO classes (set A). Finally, the mask branch is applied on all 3k Visual Genome classes (set A+B).

Before training on the mask data of the 80 COCO classes (set A) in Stage 2, a "surgery" is done to convert the 3k VG detection weights to 80 COCO detection weights, so that the mask branch only predicts mask outputs of the 80 COCO classes (as the weight transfer function only takes as input 80 classes) to save GPU memory. After training, another "surgery" is done to convert the 80 COCO detection weights back to the 3k VG detection weights.

To run the experiments, use tools/train_net.py to run the following yaml config files for each experiment with ResNet-50-FPN backbone or ResNet-101-FPN backbone.

Using ResNet-50-FPN backbone:

  1. Stage 1 (bbox training on 3k VG classes): run tools/train_net.py with configs/bbox2mask_vg/eval_sw/stage1_e2e_fast_rcnn_R-50-FPN_1x_1im.yaml
  2. Weights "surgery" 1: convert 3k VG detection weights to 80 COCO detection weights:
    python2 tools/vg3k_training/convert_coco_seg_to_vg3k.py --input_model /path/to/model_1.pkl --output_model /path/to/model_1_vg3k2coco_det.pkl
    where /path/to/model_1.pkl is the path to the final model trained in Stage 1 above.
  3. Stage 2 (mask training on 80 COCO classes): run tools/train_net.py with configs/bbox2mask_vg/eval_sw/stage2_cocomask_clsbox_2_layer_mlp_nograd.yaml
    IMPORTANT: when training Stage 2, set TRAIN.WEIGHTS to /path/to/model_1_vg3k2coco_det.pkl (the output of convert_coco_seg_to_vg3k.py) in tools/train_net.py.
  4. Weights "surgery" 2: convert 80 COCO detection weights back to 3k VG detection weights:
    python2 tools/vg3k_training/convert_vg3k_det_to_coco.py --input_model /path/to/model_2.pkl --output_model /path/to/model_2_coco2vg3k_seg.pkl
    where /path/to/model_2.pkl is the path to the final model trained in Stage 2 above. The output /path/to/model_2_coco2vg3k_seg.pkl can be used for VG 3k instance segmentation.

Using ResNet-101-FPN backbone:

  1. Stage 1 (bbox training on 3k VG classes): run tools/train_net.py with configs/bbox2mask_vg/eval_sw_R101/stage1_e2e_fast_rcnn_R-101-FPN_1x_1im.yaml
  2. Weights "surgery" 1: convert 3k VG detection weights to 80 COCO detection weights:
    python2 tools/vg3k_training/convert_coco_seg_to_vg3k.py --input_model /path/to/model_1.pkl --output_model /path/to/model_1_vg3k2coco_det.pkl
    where /path/to/model_1.pkl is the path to the final model trained in Stage 1 above.
  3. Stage 2 (mask training on 80 COCO classes): run tools/train_net.py with configs/bbox2mask_vg/eval_sw_R101/stage2_cocomask_clsbox_2_layer_mlp_nograd_R101.yaml
    IMPORTANT: when training Stage 2, set TRAIN.WEIGHTS to /path/to/model_1_vg3k2coco_det.pkl (the output of convert_coco_seg_to_vg3k.py) in tools/train_net.py.
  4. Weights "surgery" 2: convert 80 COCO detection weights back to 3k VG detection weights:
    python2 tools/vg3k_training/convert_vg3k_det_to_coco.py --input_model /path/to/model_2.pkl --output_model /path/to/model_2_coco2vg3k_seg.pkl
    where /path/to/model_2.pkl is the path to the final model trained in Stage 2 above. The output /path/to/model_2_coco2vg3k_seg.pkl can be used for VG 3k instance segmentation.

(Alternatively, you may skip Stage 1 and Weights "surgery" 1 by directly downloading the pre-trained VG 3k detection weights by running bash lib/datasets/data/trained_models/fetch_vg3k_faster_rcnn_model.sh, and leaving TRAIN.WEIGHTS to the specified values in the yaml configs in Stage 2.)

Owner
Ronghang Hu
Research Scientist, Facebook AI Research (FAIR)
Ronghang Hu
🖺 OCR using tensorflow with attention

tensorflow-ocr 🖺 OCR using tensorflow with attention, batteries included Installation git clone --recursive http://github.com/pannous/tensorflow-ocr

646 Nov 11, 2022
list all open dataset about ocr.

ocr-open-dataset list all open dataset about ocr. printed dataset year Born-Digital Images (Web and Email) 2011-2015 COCO-Text 2017 Text Extraction fr

hongbomin 95 Nov 24, 2022
Détection de créneaux de vaccination disponibles pour l'outil ViteMaDose

Vite Ma Dose ! est un outil open source de CovidTracker permettant de détecter les rendez-vous disponibles dans votre département afin de vous faire v

CovidTracker 239 Dec 13, 2022
かの有名なあの東方二次創作ソング、「bad apple!」のMVをPythonでやってみたって話

bad apple!! 内容 このプログラムは、bad apple!(feat. nomico)のPVをPythonを用いて再現しよう!という内容です。 実はYoutube並びにGithub上に似たようなプログラムがあったしなんならそっちの方が結構良かったりするんですが、一応公開しますw 使い方 こ

赤紫 8 Jan 05, 2023
The first open-source library that detects the font of a text in a image.

Typefont Typefont is an experimental library that detects the font of a text in a image. Usage Import the main function and invoke it like in the foll

Vasile Pește 1.6k Feb 24, 2022
Python tool that takes the OCR.space JSON output as input and draws a text overlay on top of the image.

OCR.space OCR Result Checker = Draw OCR overlay on top of image Python tool that takes the OCR.space JSON output as input, and draws an overlay on to

a9t9 4 Oct 18, 2022
Dirty, ugly, and hopefully useful OCR of Facebook Papers docs released by Gizmodo

Quick and Dirty OCR of Facebook Papers Gizmodo has been working through the Facebook Papers and releasing the docs that they process and review. As lu

Bill Fitzgerald 2 Oct 28, 2021
Perspective recovery of text using transformed ellipses

unproject_text Perspective recovery of text using transformed ellipses. See full writeup at https://mzucker.github.io/2016/10/11/unprojecting-text-wit

Matt Zucker 111 Nov 13, 2022
1st place solution for SIIM-FISABIO-RSNA COVID-19 Detection Challenge

SIIM-COVID19-Detection Source code of the 1st place solution for SIIM-FISABIO-RSNA COVID-19 Detection Challenge. 1.INSTALLATION Ubuntu 18.04.5 LTS CUD

Nguyen Ba Dung 170 Dec 21, 2022
With the virtual keyboard, you can write on the real time images by combining the thumb and index fingers on the letter you want.

Virtual Keyboard With the virtual keyboard, you can write on the real time images by combining the thumb and index fingers on the letter you want. At

Güldeniz Bektaş 5 Jan 23, 2022
Converts an image into funny, smaller amongus characters

SussyImage Converts an image into funny, smaller amongus characters Demo Mona Lisa | Lona Misa (Made up of AmongUs characters) API I've also added an

Dhravya Shah 14 Aug 18, 2022
question‘s area recognition using image processing and regular expression

======================================== Paper-Question-recognition ======================================== question‘s area recognition using image p

Yuta Mizuki 7 Dec 27, 2021
PyTorch Re-Implementation of EAST: An Efficient and Accurate Scene Text Detector

Description This is a PyTorch Re-Implementation of EAST: An Efficient and Accurate Scene Text Detector. Only RBOX part is implemented. Using dice loss

365 Dec 20, 2022
Handwritten Text Recognition (HTR) using TensorFlow 2.x

Handwritten Text Recognition (HTR) system implemented using TensorFlow 2.x and trained on the Bentham/IAM/Rimes/Saint Gall/Washington offline HTR data

Arthur Flôr 160 Dec 21, 2022
Bu uygulamada Python ve Opencv kullanarak bilgisayar kamerasından yüz tespiti yapıyoruz.

opencv_yuz_bulma Bu uygulamada Python ve Opencv kullanarak bilgisayar kamerasından yüz tespiti yapıyoruz. Bilgisarın kendi kamerasını kullanmak için;

Ahmet Haydar Ornek 6 Apr 16, 2022
Detect and fix skew in images containing text

Alyn Skew detection and correction in images containing text Image with skew Image after deskew Install and use via pip! Recommended way(using virtual

Kakul 230 Dec 21, 2022
Convolutional Recurrent Neural Networks(CRNN) for Scene Text Recognition

CRNN_Tensorflow This is a TensorFlow implementation of a Deep Neural Network for scene text recognition. It is mainly based on the paper "An End-to-En

MaybeShewill-CV 1000 Dec 27, 2022
Memory tests solver with using OpenCV

Human Benchmark project This project is OpenCV based programs which are puzzle solvers for 7 different games for https://humanbenchmark.com/. made as

Bahadır Araz 24 Dec 27, 2022
This repository summarized computer vision theories.

This repository summarized computer vision theories.

3 Feb 04, 2022