Repository: valeoai/LOST
Branch: master
Commit: fcedbecb644f
Files: 31
Total size: 34.6 MB
Directory structure:
gitextract_b6vyv_32/
├── .gitignore
├── LICENSE
├── README.md
├── __init__.py
├── cluster_for_OD.py
├── data/
│ ├── CAD_predictions/
│ │ ├── LOST_plus_CAD_COCO20k.json
│ │ ├── LOST_plus_CAD_VOC07.json
│ │ └── LOST_plus_CAD_VOC12.json
│ └── LOST_predictions/
│ └── LOST_VOC07.pkl
├── datasets/
│ └── coco_20k_filenames.txt
├── datasets.py
├── main_corloc_evaluation.py
├── main_lost.py
├── networks.py
├── object_discovery.py
├── requirements.txt
├── tools/
│ ├── configs/
│ │ ├── RN50_DINO_FRCNN_COCO20k_CAD.yaml
│ │ ├── RN50_DINO_FRCNN_VOC07_CAD.yaml
│ │ ├── RN50_DINO_FRCNN_VOC07_OD.yaml
│ │ ├── RN50_DINO_FRCNN_VOC12_CAD.yaml
│ │ └── RN50_DINO_FRCNN_VOC12_OD.yaml
│ ├── convert_pretrained_to_detectron_format.py
│ ├── evaluate_unsupervised_detection_voc.py
│ ├── prepare_coco_CAD_gt.py
│ ├── prepare_coco_LOST_CAD_pseudo_boxes_in_detectron2_format.py
│ ├── prepare_voc_LOST_CAD_pseudo_boxes_in_detectron2_format.py
│ ├── prepare_voc_LOST_OD_pseudo_boxes_in_detectron2_format.py
│ ├── prepare_voc_data_in_coco_style.py
│ ├── train_net_for_LOST_CAD.py
│ └── train_net_for_LOST_OD.py
└── visualizations.py
================================================
FILE CONTENTS
================================================
================================================
FILE: .gitignore
================================================
outputs/*
*.pyc
================================================
FILE: LICENSE
================================================
LOST
Copyright 2021 Valeo
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
https://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
Apache License
Version 2.0, January 2004
https://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
================================================
FILE: README.md
================================================
# LOST
Pytorch implementation of the unsupervised object discovery method **LOST**. More details can be found in the paper:
**Localizing Objects with Self-Supervised Transformers and no Labels**, BMVC 2021 [[arXiv](https://arxiv.org/abs/2109.14279)]
by *Oriane Siméoni, Gilles Puy, Huy V. Vo, Simon Roburin, Spyros Gidaris, Andrei Bursuc, Patrick Pérez, Renaud Marlet and Jean Ponce*
\
If you use the **LOST** code or framework in your research, please consider citing:
```
@inproceedings{LOST,
title = {Localizing Objects with Self-Supervised Transformers and no Labels},
author = {Oriane Sim\'eoni and Gilles Puy and Huy V. Vo and Simon Roburin and Spyros Gidaris and Andrei Bursuc and Patrick P\'erez and Renaud Marlet and Jean Ponce},
journal = {Proceedings of the British Machine Vision Conference (BMVC)},
month = {November},
year = {2021}
}
```
## Content
#### LOST
- [Installation of LOST](#installation-of-lost)
- [Apply LOST to one image](#apply-lost-to-one-image)
- [Launching LOST on datasets](#launching-lost-on-datasets)
#### Towards unsupervised object detection
- [Installation of LOST+CAD/OD](#installation-for-cad-and-od-trainings)
- [Training LOST+CAD](#training-a-class-agnostic-detector-cad-with-lost-pseudo-annotations)
- [Training LOST+OD](#training-a-class-aware-detector-od-with-lost-pseudo-annotations)
## Installation of LOST
### Dependencies
This code was implemented with python 3.7, PyTorch 1.7.1 and CUDA 10.2. Please install [PyTorch](https://pytorch.org/). In order to install the additionnal dependencies, please launch the following command:
```
pip install -r requirements.txt
```
### Install DINO
This method is based on DINO [paper](https://arxiv.org/pdf/2104.14294.pdf). The framework can be installed using the following commands:
```
git clone https://github.com/facebookresearch/dino.git
cd dino;
touch __init__.py
echo -e "import sys\nfrom os.path import dirname, join\nsys.path.insert(0, join(dirname(__file__), '.'))" >> __init__.py; cd ../;
```
The code was made using the commit ba9edd1 of DINO repo (please rebase if breakage).
## Apply LOST to one image
Following are scripts to apply LOST to an image defined via the `image_path` parameter and visualize the predictions (`pred`), the maps of the Figure 2 in the paper (`fms`) and the visulization of the seed expansion (`seed_expansion`). Box predictions are also stored in the output directory given by parameter `output_dir`.
```
python main_lost.py --image_path examples/VOC07_000236.jpg --visualize pred
python main_lost.py --image_path examples/VOC07_000236.jpg --visualize fms
python main_lost.py --image_path examples/VOC07_000236.jpg --visualize seed_expansion
```
## Launching LOST on datasets
Following are the different steps to reproduce the results of **LOST** presented in the paper.
### PASCAL-VOC
Please download the PASCAL VOC07 and PASCAL VOC12 datasets ([link](http://host.robots.ox.ac.uk/pascal/VOC/)) and put the data in the folder `datasets`. There should be the two subfolders: `datasets/VOC2007` and `datasets/VOC2012`. In order to apply lost and compute corloc results (VOC07 61.9, VOC12 64.0), please launch:
```
python main_lost.py --dataset VOC07 --set trainval
python main_lost.py --dataset VOC12 --set trainval
```
### COCO
Please download the [COCO dataset](https://cocodataset.org/#home) and put the data in `datasets/COCO`. Results are provided given the 2014 annotations following previous works. The following command line allows you to get results on the subset of 20k images of the COCO dataset (corloc 50.7), following previous litterature. To be noted that the 20k images are a subset of the `train` set.
```
python main_lost.py --dataset COCO20k --set train
```
### Different models
We have tested the method on different setups of the VIT model, corloc results are presented in the following table (more can be found in the paper).
| arch |
pre-training |
dataset |
|
|
VOC07 |
VOC12 |
COCO20k |
| ViT-S/16 |
DINO |
61.9 |
64.0 |
50.7 |
| ViT-S/8 |
DINO |
55.5 |
57.0 |
49.5 |
| ViT-B/16 |
DINO |
60.1 |
63.3 |
50.0 |
| ResNet50 |
DINO |
36.8 |
42.7 |
26.5 |
| ResNet50 |
Imagenet |
33.5 |
39.1 |
25.5 |
\
Previous results on the dataset `VOC07` can be obtained by launching:
```
python main_lost.py --dataset VOC07 --set trainval #VIT-S/16
python main_lost.py --dataset VOC07 --set trainval --patch_size 8 #VIT-S/8
python main_lost.py --dataset VOC07 --set trainval --arch vit_base #VIT-B/16
python main_lost.py --dataset VOC07 --set trainval --arch resnet50 #Resnet50/DINO
python main_lost.py --dataset VOC07 --set trainval --arch resnet50_imagenet #Resnet50/imagenet
```
## Towards unsupervised object detection
In this work, we additionally use LOST predictions to train object detection models without any human supervision. We explore two scenarios: class-agnostic (CAD) and (pseudo) class-aware training of object detectors (OD). The next section present the different steps to reproduce our results.
### Installation for CAD and OD trainings
We use the [detectron2](https://github.com/facebookresearch/detectron2) framework to train a Faster R-CNN model with LOST predictions as pseudo-gt. The code was developped with the version [v0.5](https://github.com/facebookresearch/detectron2/releases) of the framework. In order to reproduce our results, please install detectron2 using the next commands. In case of failure, you can find the installation corresponding to your version of pytorch/CUDA [here](https://github.com/facebookresearch/detectron2/releases).
```bash
git clone https://github.com/facebookresearch/detectron2.git
python -m pip install detectron2==0.5
```
Set global variables for ease of usage.
```bash
export LOST=$(pwd)
cd detectron2; export D2=$(pwd);
```
Then please copy LOST-specific files to detectron2 framework, following:
```bash
ln -s $LOST/tools/*.py $D2/tools/. # Move LOST tools to D2
mkdir $D2/configs/LOST
ln -s $LOST/tools/configs/* $D2/configs/LOST/. # Move LOST configs to D2
```
### Training a Class-Agnostic Detector (CAD) with LOST pseudo-annotations
Before launching a training, data must be formated to fit detectron2 and COCO styles. Following are the command lines to do this formatting for boxes predicted with LOST.
```bash
cd $D2;
# Format DINO weights to fit detectron2
wget https://dl.fbaipublicfiles.com/dino/dino_resnet50_pretrain/dino_resnet50_pretrain.pth -P ./data # Download the model from DINO
python tools/convert_pretrained_to_detectron_format.py --input ./data/dino_resnet50_pretrain.pth --output ./data/dino_RN50_pretrain_d2_format.pkl
# Format pseudo-boxes data to fit detectron2
python tools/prepare_voc_LOST_CAD_pseudo_boxes_in_detectron2_format.py --year 2007 --pboxes $LOST/data/LOST_predictions/LOST_VOC07.pkl
# Format VOC data to fit COCO style
python tools/prepare_voc_data_in_coco_style.py --is_CAD --voc07_dir $LOST/datasets/VOC2007 --voc12_dir $LOST/datasets/VOC2012
```
The next command line allows you to launch a CAD training with 4 gpus on the VOC2007 dataset. The batch size is set to 16, 4 to 8 GPUs may be needed depending on your machines. Please make sure to change the argument value `MODEL.WEIGHTS` to the correct path of DINO weights.
```bash
python tools/train_net_for_LOST_CAD.py --num-gpus 4 --config-file ./configs/LOST/RN50_DINO_FRCNN_VOC07_CAD.yaml DATALOADER.NUM_WORKERS 8 OUTPUT_DIR ./outputs/RN50_DINO_FRCNN_VOC07_CAD MODEL.WEIGHTS ./data/dino_RN50_pretrain_d2_format.pkl
```
Inference results of the model will be stored in `$OUTPUT_DIR/inference`. In order to produce results on the `train+val` dataset, please use the following command:
```
python tools/train_net_for_LOST_CAD.py --resume --eval-only --num-gpus 4 --config-file ./configs/LOST/RN50_DINO_FRCNN_VOC07_CAD.yaml DATALOADER.NUM_WORKERS 6 MODEL.WEIGHTS ./outputs/RN50_DINO_FRCNN_VOC07_CAD/model_final.pth OUTPUT_DIR ./outputs/RN50_DINO_FRCNN_VOC07_CAD/ DATASETS.TEST '("voc_2007_trainval_CAD_coco_style", )'
cd $LOST;
python main_corloc_evaluation.py --dataset VOC07 --set trainval --type_pred detectron --pred_file $D2/outputs/RN50_DINO_FRCNN_VOC07_CAD/inference/coco_instances_results.json
```
#### Training LOST+CAD on COCO20k dataset
Following are the command lines allowing to train a detector in a class-agnostic fashion on the COCO20k subset of COCO dataset.
```bash
cd $D2;
# Format pseudo-boxes data to fit detectron2
python tools/prepare_coco_LOST_CAD_pseudo_boxes_in_detectron2_format.py --pboxes $LOST/outputs/COCO20k_train/LOST-vit_small16_k/preds.pkl
# Generate COCO20k CAD gt annotations
python tools/prepare_coco_CAD_gt.py --coco_dir $LOST/datasets/COCO
# Train detector (evaluation done on COCO20k CAD training set)
python tools/train_net_for_LOST_CAD.py --num-gpus 4 --config-file ./configs/LOST/RN50_DINO_FRCNN_COCO20k_CAD.yaml DATALOADER.NUM_WORKERS 8 OUTPUT_DIR ./outputs/RN50_DINO_FRCNN_COCO20k_CAD MODEL.WEIGHTS ./data/dino_RN50_pretrain_d2_format.pkl
# Corloc evaluation
python main_corloc_evaluation.py --dataset COCO20k --type_pred detectron --pred_file $D2/outputs/RN50_DINO_FRCNN_COCO20k_CAD/inference/coco_instances_results.json
```
#### Evaluating LOST+CAD (corloc results)
We have provided predictions of a class-agnostic Faster R-CNN model trained using LOST boxes as pseudo-gt; they are stored in the folder `data/CAD_predictions`. In order to launch the corloc evaluation, please launch the following scripts. It is to be noted that in this evaluation, only the box with the highest confidence score is considered per image.
```bash
python main_corloc_evaluation.py --dataset VOC07 --set trainval --type_pred detectron --pred_file data/CAD_predictions/LOST_plus_CAD_VOC07.json
python main_corloc_evaluation.py --dataset VOC12 --set trainval --type_pred detectron --pred_file data/CAD_predictions/LOST_plus_CAD_VOC12.json
python main_corloc_evaluation.py --dataset COCO20k --set train --type_pred detectron --pred_file data/CAD_predictions/LOST_plus_CAD_COCO20k.json
```
The following table presents the obtained corloc results.
| method |
dataset |
|
VOC07 |
VOC12 |
COCO20k |
| LOST |
61.9 |
64.0 |
50.7 |
| LOST+CAD |
65.7 |
70.4 |
57.5 |
### Training a Class-Aware Detector (OD) with LOST pseudo-annotations
Following are the different steps to train a class-aware detector using LOST peusdo-boxes for the dataset VOC07. We provide LOST boxes correspoding to the dataset VOC07 in `$LOST/data/LOST_predictions/LOST_VOC07.pkl`.
```bash
cd $LOST;
# Cluster features of LOST boxes
python cluster_for_OD.py --pred_file $LOST/data/LOST_predictions/LOST_VOC07.pkl --nb_clusters 20 --dataset VOC07 --set trainval
cd $D2;
# Format DINO weights to fit detectron2
wget https://dl.fbaipublicfiles.com/dino/dino_resnet50_pretrain/dino_resnet50_pretrain.pth -P ./data # Download the model from DINO
python tools/convert_pretrained_to_detectron_format.py --input ./data/dino_resnet50_pretrain.pth --output ./data/dino_RN50_pretrain_d2_format.pkl
# Prepare the clustered LOST pseudo-box data for training
python tools/prepare_voc_LOST_OD_pseudo_boxes_in_detectron2_format.py --year 2007 --pboxes $LOST/data/LOST_predictions/LOST_VOC07_clustered_20clu.pkl
# Format VOC data to fit COCO style
python tools/prepare_voc_data_in_coco_style.py --voc07_dir $LOST/datasets/VOC2007 --voc12_dir $LOST/datasets/VOC2012
# Train the detector on VOC2007 trainval set -- please be aware that no hungarian matching is used during training, so validation restuls are not meaningful (will be close to 0). Please use command bellow in order to evaluate results correctly.
python tools/train_net_for_LOST_OD.py --num-gpus 8 --config-file ./configs/LOST/RN50_DINO_FRCNN_VOC07_OD.yaml DATALOADER.NUM_WORKERS 8 OUTPUT_DIR ./outputs/RN50_DINO_FRCNN_VOC07_OD MODEL.WEIGHTS ./data/dino_RN50_pretrain_d2_format.pkl
# Evaluate the detector results using hungarian matching -- allows to reproduce results from the paper
cd $LOST;
python tools/evaluate_unsupervised_detection_voc.py --results ./detectron2/outputs/RN50_DINO_FRCNN_VOC07_OD/inference/coco_instances_results.json
```
### Training details
We use the `R50-C4` model of Detectron2 with ResNet50 pre-trained with DINO self-supervision [model](https://dl.fbaipublicfiles.com/dino/dino_resnet50_pretrain/dino_resnet50_pretrain.pth).
Details:
- mini-batches of size 16 across 8 GPUs using SyncBatchNorm
- extra BatchNorm layer for the RoI head after conv5, i.e., `Res5ROIHeadsExtraNorm` layer in Detectron2
- frozen first two convolutional blocks of ResNet-50, i.e., `conv1` and `conv2` in Detectron2.
- learning rate is first warmed-up for 100 steps to 0.02 and then reduced by a factor of 10 after 18K and 22K training steps
- we use in total 24K training steps for all the experiments, except when training class-agnostic detectors on
the pseudo-boxes of the VOC07 trainval set, in which case we use 10K steps.
## License
LOST is released under the [Apache 2.0 license](./LICENSE).
================================================
FILE: __init__.py
================================================
================================================
FILE: cluster_for_OD.py
================================================
# Copyright 2021 - Valeo Comfort and Driving Assistance - Oriane Siméoni @ valeo.ai
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import os
import argparse
import pickle
from tqdm import tqdm
from collections import defaultdict
import torch
import torch.nn as nn
import torchvision
from torchvision import transforms as pth_transforms
import numpy as np
import scipy.cluster.vq as vq
from networks import get_model
from datasets import Dataset
if __name__ == "__main__":
parser = argparse.ArgumentParser("Cluster LOST predictions.")
# Model
parser.add_argument(
"--arch",
default="vit_small",
type=str,
choices=[
"vit_small",
],
help="Model architecture.",
)
parser.add_argument(
"--patch_size",
default=16,
type=int,
help="Patch resolution of the model."
)
# Dataset
parser.add_argument(
"--dataset",
default="VOC07",
type=str,
choices=[None, "VOC07", "VOC12", "COCO20k"],
help="Dataset name.",
)
parser.add_argument(
"--set",
default="train",
type=str,
choices=["val", "train", "trainval", "test"],
help="Path of the image to load.",
)
parser.add_argument(
"--no_hard",
action="store_true",
help="Only used in the case of the VOC_all setup (see the paper).",
)
# Prediction files
parser.add_argument(
"--pred_file",
type=str,
default="outputs/VOC07_trainval/LOST-vit_small16_k/preds.pkl",
help="Predicted boxes.",
)
# Clustering specific
parser.add_argument(
"--nb_clusters",
type=int,
default=20,
help="Number of clusters used for kmeans clustering.")
parser.add_argument("--random_seed",
type=int,
default=123,
help="K-means random seed.")
# Keep?
parser.add_argument("--visualize", type=str, default=None, help="Visualize")
args = parser.parse_args()
# -------------------------------------------------------------------------------------------------------
# Dataset
dataset = Dataset(args.dataset, args.set, args.no_hard)
# -------------------------------------------------------------------------------------------------------
# Model
device = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu")
model = get_model(args.arch, args.patch_size, -1, device)
# -------------------------------------------------------------------------------------------------------
# Load predictions
print(f'Extract features corresponding to the boxes {args.pred_file}.')
with open(args.pred_file, "rb") as f:
predictions = pickle.load(f)
# -------------------------------------------------------------------------------------------------------
# Extract CLS token
# Features location
out_path = f'{args.pred_file.split(".pkl")[0]}_cropped_feats_{args.arch}.pkl'
if not os.path.exists(out_path):
feats = defaultdict(defaultdict)
pbar = tqdm(dataset.dataloader)
for im_id, inp in enumerate(pbar):
# ------------ Image processing ---------------------------------------
img = inp[0]
init_image_size = img.shape
# Get the name of the image
im_name = dataset.get_image_name(inp[1])
# Pass in case of no gt boxes in the image
if im_name is None:
continue
# Prediction
pred = np.asarray(predictions[im_name])
xmin, xmax = round(pred[1]), round(pred[3])
ymin, ymax = round(pred[0]), round(pred[2])
# Crop the image
cropped = img[:, xmin:xmax, ymin:ymax]
# Resize cropped region
resize_f = pth_transforms.Resize(256, interpolation=3)
cropped_im = resize_f(cropped)
# move to gpu
cropped_im = cropped_im.cuda(non_blocking=True)
# Size for transformers
w_featmap = cropped_im.shape[-2] // args.patch_size
h_featmap = cropped_im.shape[-1] // args.patch_size
# Forward pass
with torch.no_grad():
f = model(cropped_im[None, :, :, :])
norm_f = nn.functional.normalize(f, dim=1, p=2)
feats[im_name]["cropped_feat"] = np.array(norm_f.to("cpu"))
feats[im_name]["predicted_bb"] = predictions[im_name]
with open(out_path, "wb") as handle:
pickle.dump(feats, handle, protocol=pickle.HIGHEST_PROTOCOL)
print(f'Cropped features saved at {out_path}.')
else:
with open(out_path, "rb") as f:
feats = pickle.load(f)
print(f'Cropped features loaded from {out_path}.')
# -------------------------------------------------------------------------------------------------------
# Apply clustering
seed_ = f'_seed-{args.random_seed}' if args.random_seed != 123 else ""
clustering_path = f'{args.pred_file.split(".pkl")[0]}_clustered_{args.nb_clusters}clu{seed_}.pkl'
np.random.seed(seed=args.random_seed)
all_feats = []
pred_bbx = []
keys = sorted(feats.keys())
for key in keys:
if feats[key]["cropped_feat"].squeeze().shape == (384,):
all_feats.append(feats[key]["cropped_feat"].squeeze())
pred_bbx.append(feats[key]["predicted_bb"])
# Cluster whitened features
x = np.array(all_feats)
c, clusters = vq.kmeans2(data=vq.whiten(x) / np.linalg.norm(vq.whiten(x), axis=1)[:, None],
k=args.nb_clusters)
pseudo_labels = defaultdict(defaultdict)
for i in range(len(keys)):
k = keys[i]
pseudo_labels[k]["pseudo_label"] = clusters[i]
pseudo_labels[k]["predicted_bb"] = pred_bbx[i]
with open(clustering_path, "wb") as f:
pickle.dump(pseudo_labels, f, protocol=pickle.HIGHEST_PROTOCOL)
