Home

Awesome

SUMMIT: Source-Free Adaptation of Uni-Modal Models to Multi-Modal Targets

Official code for the paper.

Paper

SUMMIt: Source-Free Adaptation fo Uni-Modal Models to Multi-Modal Targets
Cody Simons, Dripta Raychaudhuri, Sk Miraj Ahmed, Suya You, Konstantinos Karydis, Amit K. Roy-Chowdhury University of California, Riverside & Army Research Lab ICCV 2023

If you find this code useful for your research, please cite our paper:

@inproceedings{simons2023summit,
    title={SUMMIT: Source-Free Adaptationof Uni-Modal Models to Multi-Modal Targets},
    author={Simons, Cody and Raychaudhuri, Dripta and Ahmed, Sk Miraj and Karydis, Konstantinos and You, Suya and Roy-Chowdhury, Amit},
    booktitle={ICCV},
    year={2023}
}

Preparation

Prerequisites

Tested with

Installation

For installation please follow all instructions to install xMUDA

Datasets

Please refer to xMUDA for instructions on downloading the NuScenes, A2D2, and SemanticKITTI datasets.

NuScenes LidarSeg

Please download the Full dataset (v1.0) from the NuScenes website and extract it.

You need to perform preprocessing to generate the data for xMUDA first. The preprocessing subsamples the 360° LiDAR point cloud to only keep the points that project into the front camera image. It also generates the point-wise segmentation labels using the 3D objects by checking which points lie inside the 3D boxes. All information will be stored in a pickle file (except the images which will be read frame by frame by the dataloader during training).

Please edit the script xmuda/data/nuscenes/preprocess.py as follows and then run it.

Training

Baseline

Train the baselines (only on source) with:

$ python xmuda/train_baseline.py --cfg=configs/nuscenes/usa_singapore/baseline.yaml
$ python xmuda/train_baseline.py --cfg=configs/nuscenes/day_night/baseline.yaml
$ python xmuda/train_baseline.py --cfg=configs/a2d2_semantic_kitti/baseline.yaml

Pseudo-Label Generation

After having trained the xMUDA model, generate the pseudo-labels as follows:

$ python xmuda/test.py --cfg=configs/nuscenes/usa_singapore/xmuda.yaml --pselab @/model_2d_100000.pth @/model_3d_100000.pth DATASET_TARGET.TEST "('train_singapore',)"

Note that we use the last model at 100,000 steps to exclude supervision from the validation set by picking the best weights. The pseudo labels and maximum probabilities are saved as .npy file.

Pseudo-labels are then further refined using a script as follows.

$ python xmuda/refine_pseudo_labels.py /path/to/pseudo/labels /output/path --AF
$ python xmuda/refine_pseudo_labels.py /path/to/pseudo/labels /output/path --EW --HT --k 0.5

Please edit the pselab_paths in the config file, e.g. configs/nuscenes/usa_singapore/xmuda_pl_SF.yaml, to match your path of the refined pseudo-lables.

SUMMIT

You can run the training with

$ python xmuda/train_xmuda_SF.py --cfg=configs/nuscenes/usa_singapore/xmuda_pl_SF.yaml

The output will be written to /home/<user>/workspace/outputs/xmuda/<config_path> by default. The OUTPUT_DIR can be modified in the config file in (e.g. configs/nuscenes/usa_singapore/xmuda.yaml) or optionally at run time in the command line (dominates over config file). Note that @ in the following example will be automatically replaced with the config path, i.e. with nuscenes/usa_singapore/xmuda.

$ python xmuda/train_xmuda_SF.py --cfg=configs/nuscenes/usa_singapore/xmuda_pl_SF.yaml OUTPUT_DIR path/to/output/directory/@

You can start the trainings on the other UDA scenarios (Day/Night and A2D2/SemanticKITTI) analogously:

$ python xmuda/train_xmuda_SF.py --cfg=configs/nuscenes/day_night/xmuda_pl_SF.yaml
$ python xmuda/train_xmuda_SF.py --cfg=configs/a2d2_semantic_kitti/xmuda_pl_SF.yaml

Testing

You can provide which checkpoints you want to use for testing. We used the ones that performed best on the validation set during training (the best val iteration for 2D and 3D is shown at the end of each training). Note that @ will be replaced by the output directory for that config file. For example:

$ cd <root dir of this repo>
$ python xmuda/test.py --cfg=configs/nuscenes/usa_singapore/xmuda_pl_SF.yaml @/model_2d_065000.pth @/model_3d_095000.pth

You can also provide an absolute path without @.

Acknowledgements

Note that this code builds on the xMUDA repo.

License

SUMMIT is released under the Apache 2.0 license.