Awesome

MIANet

Official PyTorch Implementation of MIANet: Aggregating Unbiased Instance and General Information for Few-Shot Semantic Segmentation(CVPR 2023). This paper can be downloaded from here. <img src="/figure/logo.png" width="1000" height="250" align="bottom" />

Abstract: Existing few-shot segmentation methods are based on the meta-learning strategy and extract instance knowledge from a support set and then apply the knowledge to segment target objects in a query set. However, the extracted knowledge is insufficient to cope with the variable intraclass differences since the knowledge is obtained from a few samples in the support set. To address the problem, we propose a multi-information aggregation network (MIANet) that effectively leverages the general knowledge, i.e., semantic word embeddings, and instance information for accurate segmentation. Specifically, in MIANet, a general information module (GIM) is proposed to extract a general class prototype from word embeddings as a supplement to instance information. To this end, we design a triplet loss that treats the general class prototype as an anchor and samples positive-negative pairs from local features in the support set. The calculated triplet loss can transfer semantic similarities among language identities from a word embedding space to a visual representation space. To alleviate the model biasing towards the seen training classes and to obtain multi-scale information, we then introduce a non-parametric hierarchical prior module (HPM) to generate unbiased instance-level information via calculating the pixel-level similarity between the support and query image features. Finally, an information fusion module (IFM) combines the general and instance information to make predictions for the query image. Extensive experiments on PASCAL-5i and COCO-20i show that MIANet yields superior performance and set a new state-of-the-art.

pipeiline

🔧Get Started

Just follow these steps to train and test MIANet.

Dataset (Following the BAM operations)

1. Download the dataset from the following links.

PASCAL-5i: PASCAL VOC 2012 + SBD
COCO-20i: MSCOCO2014

2. Adjust these files to the following directory:

                                             JPEGImages
                 VOCdevkit2012--> VOC2012-->
                                             SegmentationClassAug
  MIANet/data---->           
                                annotations
                 MSCOCO2014-->  train2014  
                                val2014

3. Download the data lists (.txt files) and put them into the MIANet/lists_bam directory.

Models

(Pretrained PSPNet) Download the pre-trained PSPNet from here and put them under initmodel/pascal and initmodel/coco directory.

Scripts

1. Change configuration via the .yaml files in MIANet/config.
2. Train MIANet under the settings (resnet50 backbone; fold=0; shot=1): run train.py.
3. Test MIANet (fill the weight parameter in .yaml file): run test.py

Visualization

result

To-Do List

Complete the MIANet repository.

References

This repo is mainly built based on PFENet, BAM. Thanks for their great work!