Learning to Segment the Tail

[arXiv]

In this repository, we release code for Learning to Segment The Tail (LST). The code is directly modified from the project maskrcnn_benchmark, which is an excellent codebase! If you get any problem that causes you unable to run the project, you can check the issues under maskrcnn_benchmark first.

Installation

Please following INSTALL.md for maskrcnn_benchmark. For experiments on LVIS_v0.5 dataset, you need to use lvis-api.

LVIS Dataset

After downloading LVIS_v0.5 dataset (the images are the same as COCO 2017 version), we recommend to symlink the path to the lvis dataset to datasets/ as follows

# symlink the lvis dataset
cd ~/github/LST_LVIS
mkdir -p datasets/lvis
ln -s /path_to_lvis_dataset/annotations datasets/lvis/annotations
ln -s /path_to_coco_dataset/images datasets/lvis/images

A detailed visualization demo for LVIS is LVIS_visualization. You'll find it is the most useful thing you can get from this repo :P

Dataset Pre-processing and Indices Generation

dataset_preprocess.ipynb: LVIS dataset is split into the base set and sets for the incremental phases.

balanced_replay.ipynb: We generate indices to load the LVIS dataset offline using the balanced replay scheme discussed in our paper.

Training

Our pre-trained model is model. You can trim the model and load it for LVIS training as in trim_model. Modifications to the backbone follows Mask^X R-CNN. You can also check our paper for detail.

training for base

The base training is the same as conventional training. For example, to train a model with 8 GPUs you can run:

python -m torch.distributed.launch --nproc_per_node=8 /path_to_maskrcnn_benchmark/tools/train_net.py --use-tensorboard --config-file "/path/to/config/train_file.yaml"  MODEL.RPN.FPN_POST_NMS_TOP_N_TRAIN 1000

The details about MODEL.RPN.FPN_POST_NMS_TOP_N_TRAIN is discussed in maskrcnn-benchmark.

Edit this line to initialze the dataloader with corresponding sorted category ids.

training for incremental steps

The training for each incremental phase is armed with our data balanced replay. It needs to be initialized properly here, providing the corresponding external img-id/cls-id pairs for data-loading.

get distillation

We use ground truth bounding boxes to get prediction logits using the model trained from last step. Change this to decide which classes to be distilled.

Here is an example for running:

python ./tools/train_net.py --use-tensorboard --config-file "/path/to/config/get_distillation_file.yaml" MODEL.RPN.FPN_POST_NMS_TOP_N_TRAIN 1000

The output distillation logits are saved in json format.

Evaluation

The evaluation for LVIS is a little bit different from COCO since it is not exhausted annotated, which is discussed in detail in Gupta et al.'s work.

We also report the AP for each phase and each class, which can provide better analysis.

You can run:

export NGPUS=8
python -m torch.distributed.launch --nproc_per_node=$NGPUS /path_to_maskrcnn_benchmark/tools/test_net.py --config-file "/path/to/config/train_file.yaml"

We also provide periodically testing to check the result better, as discussed in this issue.

Thanks for all the previous work and the sharing of their codes. Sorry for my ugly code and I appreciate your advice.

JoyHuYY1412/LST_LVIS