Task-Aware Asynchronous MTL with Class Incremental Contrastive Learning for Surgical Scene Understanding
To be added
To be added
To be added
DGL is a Python package dedicated to deep learning on graphs, built atop existing tensor DL frameworks (e.g. Pytorch, MXNet) and simplifying the implementation of graph-based neural networks.
- Python 3.6
- Pytorch 1.1.0
- DGL 0.3
- CUDA 10.0
- Ubuntu 16.04
In this project, we implement our method using the Pytorch and DGL library and there are three main folders:
data/
: .datasets/
: Contains the dataset needed to train the network.checkpoints/
: Conatins trained weightsevaluation/
: Contains utility tools for evaluation.Feature_extractor/
: Used to extract features from dataset images to train the scene graph and caption network for algorithm 1, stage 3 & 4.models/
: Contains network models.utils/
: Contains utility tools used for training and evaluation.
-
Task-Aware MTL Optimization and fine-tuning (Algorithm 1)
- MTL:
- TA_aption_train.py (stage 3) (pending code documentation)
- TA_scene_graph_train.py (stage 4) (pending code documentation)
- TA_MTL_finetune_train.py (stage 5) (Pending code documentation)
- MTL:
-
Vanilla MTL (MTL-V)
- MTL_V_train.py (pending code documentation)
-
Knowledge Distillation-based MTL Optmization (Algorithm 2)
- MTL_KD_train.py (pending code documentation)
-
Knowledge Distillation-based MTL Optmization and fine-tuning
- MTL_KD_train.py (pending code documentation)
- MTL_KD_train.py (pending code documentation)
- Download validation datasetTo_be_released and palce them inside the folder
datasets/
- MTL_evaluation.py
- Download checkpoints To_be_released and place them inside the folder
checkpoints/
- MTL_evaluation.py
- Set mtl_version, adapt_type and domain
Code adopted and modified from :
- Visual-Semantic Graph Attention Network for Human-Object Interaction Detecion
- Paper Visual-Semantic Graph Attention Network for Human-Object Interaction Detecion.
- Official Pytorch implementation code.
- End-to-End Incremental Learning
- Paper End-to-End Incremental Learning.
- Pytorch implementation code.
- Curriculum by smoothing
- Paper Curriculum by smoothing.
- Pytorch implementation code.