Project Page | Paper | ArXiv | Video
Qiao Gu*, Ali Kuwajerwala*, Sacha Morin*, Krishna Murthy Jatavallabhula*, Bipasha Sen, Aditya Agarwal, Corban Rivera, William Paul, Kirsty Ellis, Rama Chellappa, Chuang Gan, Celso Miguel de Melo, Joshua B. Tenenbaum, Antonio Torralba, Florian Shkurti, Liam Paull
- The codebase has been significantly refactored in the
ali-dev
branch, which provides a real-time, streamlined re-implementation that supports RGB-D video from iPhone and has a better visualization using Rerun.io. We also provide this getting started video tutorial. Please check it out! - The code for real-world mapping and navigation using a Jackal robot is open-sourced here.
- The code for localization and mapping in AI2Thor is released in the codebase. See here for instructions.
The env variables needed can be found in env_vars.bash.template
. When following the setup guide below, you can duplicate that files and change the variables accordingly for easy setup.
We recommend setting up a virtual environment using virtualenv or conda. Our code has been tested with Python 3.10.12. It may also work with other later versions. We also provide the environment.yml
file for Conda users. In generaly, directly installing conda env using .yml
file may cause some unexpected issues, so we recommand setting up the environment by the following instructions and only using the .yml
file as a reference.
Sample instructions for conda
users.
conda create -n conceptgraph anaconda python=3.10
conda activate conceptgraph
# Install the required libraries
pip install tyro open_clip_torch wandb h5py openai hydra-core distinctipy
# for yolo
pip install ultralytics
# Install the Faiss library (CPU version should be fine)
conda install -c pytorch faiss-cpu=1.7.4 mkl=2021 blas=1.0=mkl
##### Install Pytorch according to your own setup #####
# For example, if you have a GPU with CUDA 11.8 (We tested it Pytorch 2.0.1)
conda install pytorch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2 pytorch-cuda=11.8 -c pytorch -c nvidia
# Install Pytorch3D (https://github.com/facebookresearch/pytorch3d/blob/main/INSTALL.md)
# conda install pytorch3d -c pytorch3d # This detects a conflict. You can use the command below, maybe with a different version
conda install https://anaconda.org/pytorch3d/pytorch3d/0.7.4/download/linux-64/pytorch3d-0.7.4-py310_cu118_pyt201.tar.bz2
# Install the gradslam package and its dependencies
# Please clone and install them in separate folders, not within the concept-graphs folder.
git clone https://github.com/krrish94/chamferdist.git
cd chamferdist
pip install .
cd ..
git clone https://github.com/gradslam/gradslam.git
cd gradslam
git checkout conceptfusion
pip install .
Install Grounded-SAM package
Follow the instructions on the original repo. ConceptGraphs has been tested with the codebase at this commit. Grounded-SAM codebase at later commits may require some adaptations.
First checkout the package by
git clone [email protected]:IDEA-Research/Grounded-Segment-Anything.git
Then, install the package Following the commands listed in the original GitHub repo. You can skip the Install osx
step and the "optional dependencies".
During this process, you will need to set the CUDA_HOME
to be where the CUDA toolkit is installed.
The CUDA tookit can be set up system-wide or within a conda environment. We tested it within a conda environment, i.e. installing cudatoolkit-dev using conda.
# i.e. You can install cuda toolkit using conda
conda install -c conda-forge cudatoolkit-dev
# and you need to replace `export CUDA_HOME=/path/to/cuda-11.3/` by
export CUDA_HOME=/path/to/anaconda3/envs/conceptgraph/
You also need to download ram_swin_large_14m.pth
, groundingdino_swint_ogc.pth
, sam_vit_h_4b8939.pth
(and optionally tag2text_swin_14m.pth
if you want to try Tag2Text) following the instruction here.
After installation, set the path to Grounded-SAM as an environment variable
export GSA_PATH=/path/to/Grounded-Segment-Anything
Follow the installation instructions on this page. The major steps are:
- Install FastSAM codebase following here. You don't have to create a new conda env. Just installing it in the same env as the Grounded-SAM is fine.
- Download FastSAM checkpoints FastSAM-x.pt and save it to
Grounded-Segment-Anything/EfficientSAM
. - Download MobileSAM checkpoints mobile_sam.pt and save it to
Grounded-Segment-Anything/EfficientSAM
. - Download Light HQ-SAM checkpoints sam_hq_vit_tiny.pth and save it to
Grounded-Segment-Anything/EfficientSAM
.
git clone [email protected]:concept-graphs/concept-graphs.git
cd concept-graphs
pip install -e .
Follow the instructions on the LLaVA repo to set it up. You also need to prepare the LLaVA checkpoints and save them to $LLAVA_CKPT_PATH
. We have tested with model checkpoint LLaVA-7B-v0
and LLaVA code at this commit. LLaVA codebase at later commits may require some adaptations.
# Set the env variables as follows (change the paths accordingly)
export LLAVA_PYTHON_PATH=/path/to/llava
export LLAVA_CKPT_PATH=/path/to/LLaVA-7B-v0
ConceptGraphs takes posed RGB-D images as input. Here we show how to prepare the dataset using Replica as an example. Instead of the original Replica dataset, download the scanned RGB-D trajectories of the Replica dataset provided by Nice-SLAM. It contains rendered trajectories using the mesh models provided by the original Replica datasets.
Download the Replica RGB-D scan dataset using the downloading script in Nice-SLAM and set $REPLICA_ROOT
to its saved path.
export REPLICA_ROOT=/path/to/Replica
export CG_FOLDER=/path/to/concept-graphs/
export REPLICA_CONFIG_PATH=${CG_FOLDER}/conceptgraph/dataset/dataconfigs/replica/replica.yaml
ConceptGraphs can also be easily run on other dataset. See dataset/datasets_common.py
for how to write your own dataloader.
The following commands should be run in the conceptgraph
folder.
cd conceptgraph
The following command runs a 3D RGB reconstruction (GradSLAM) of a replica scene and also visualize it. This is useful for sanity check.
--visualize
requires it to be run with GUI.
SCENE_NAME=room0
python scripts/run_slam_rgb.py \
--dataset_root $REPLICA_ROOT \
--dataset_config $REPLICA_CONFIG_PATH \
--scene_id $SCENE_NAME \
--image_height 480 \
--image_width 640 \
--stride 5 \
--visualize
First, (Detection) Segmentation results and per-region CLIP features are extracted. In the following, we provide two options.
- The first one (ConceptGraphs) uses SAM in the "segment all" mode and extract class-agnostic masks.
- The second one (ConceptGraphs-Detect) uses a tagging model and a detection model to extract class-aware bounding boxes first, and then use them as prompts for SAM to segment each object.
SCENE_NAME=room0
# The CoceptGraphs (without open-vocab detector)
python scripts/generate_gsa_results.py \
--dataset_root $REPLICA_ROOT \
--dataset_config $REPLICA_CONFIG_PATH \
--scene_id $SCENE_NAME \
--class_set none \
--stride 5
# The ConceptGraphs-Detect
CLASS_SET=ram
python scripts/generate_gsa_results.py \
--dataset_root $REPLICA_ROOT \
--dataset_config $REPLICA_CONFIG_PATH \
--scene_id $SCENE_NAME \
--class_set $CLASS_SET \
--box_threshold 0.2 \
--text_threshold 0.2 \
--stride 5 \
--add_bg_classes \
--accumu_classes \
--exp_suffix withbg_allclasses
The above commands will save the detection and segmentation results in $REPLICA_ROOT/$SCENE_NAME/
.
The visualization of the detection and segmentation can be viewed in $REPLICA_ROOT/$SCENE_NAME/gsa_vis_none
and $REPLICA_ROOT/$SCENE_NAME/gsa_vis_ram_withbg_allclasses
respectively.
You can ignore the There's a wrong phrase happen, this is because of our post-process merged wrong tokens, which will be modified in the future. We will assign it with a random label at this time.
message for now.
The following command builds an object-based 3D map of the scene, using the image segmentation results from above.
- Use
save_objects_all_frames=True
to save the mapping results at every frame, which can be used for animated visualization byscripts/animate_mapping_interactive.py
andscripts/animate_mapping_save.py
. - Use
merge_interval=20 merge_visual_sim_thresh=0.8 merge_text_sim_thresh=0.8
to also perform overlap-based merging during the mapping process.
# Using the CoceptGraphs (without open-vocab detector)
THRESHOLD=1.2
python slam/cfslam_pipeline_batch.py \
dataset_root=$REPLICA_ROOT \
dataset_config=$REPLICA_CONFIG_PATH \
stride=5 \
scene_id=$SCENE_NAME \
spatial_sim_type=overlap \
mask_conf_threshold=0.95 \
match_method=sim_sum \
sim_threshold=${THRESHOLD} \
dbscan_eps=0.1 \
gsa_variant=none \
class_agnostic=True \
skip_bg=True \
max_bbox_area_ratio=0.5 \
save_suffix=overlap_maskconf0.95_simsum${THRESHOLD}_dbscan.1_merge20_masksub \
merge_interval=20 \
merge_visual_sim_thresh=0.8 \
merge_text_sim_thresh=0.8
# On the ConceptGraphs-Detect
SCENE_NAME=room0
THRESHOLD=1.2
python slam/cfslam_pipeline_batch.py \
dataset_root=$REPLICA_ROOT \
dataset_config=$REPLICA_CONFIG_PATH \
stride=5 \
scene_id=$SCENE_NAME \
spatial_sim_type=overlap \
mask_conf_threshold=0.25 \
match_method=sim_sum \
sim_threshold=${THRESHOLD} \
dbscan_eps=0.1 \
gsa_variant=ram_withbg_allclasses \
skip_bg=False \
max_bbox_area_ratio=0.5 \
save_suffix=overlap_maskconf0.25_simsum${THRESHOLD}_dbscan.1
The above commands will save the mapping results in $REPLICA_ROOT/$SCENE_NAME/pcd_saves
. It will create two pkl.gz
files, where the one with _post
suffix indicates results after some post processing, which we recommend using.`
If you run the above command with save_objects_all_frames=True
, it will create a folder in $REPLICA_ROOT/$SCENE_NAME/objects_all_frames
. Then you can run the following command to visualize the mapping process or save it to a video. Also see the relevant files for available key callbacks for viusalization options.
python scripts/animate_mapping_interactive.py --input_folder $REPLICA_ROOT/$SCENE_NAME/objects_all_frames/<folder_name>
python scripts/animate_mapping_save.py --input_folder $REPLICA_ROOT/$SCENE_NAME/objects_all_frames/<folder_name>
python scripts/visualize_cfslam_results.py --result_path /path/to/output.pkl.gz
Then in the open3d visualizer window, you can use the following key callbacks to change the visualization.
- Press
b
to toggle the background point clouds (wall, floor, ceiling, etc.). Only works on the ConceptGraphs-Detect. - Press
c
to color the point clouds by the object class from the tagging model. Only works on the ConceptGraphs-Detect. - Press
r
to color the point clouds by RGB. - Press
f
and type text in the terminal, and the point cloud will be colored by the CLIP similarity with the input text. - Press
i
to color the point clouds by object instance ID.
First, download the GT point cloud with per-point semantic segmentation labels from this Google Drive link. Please refer to this issue for a brief description of how they are generated. Unzip the file and record its location in REPLICA_SEMANTIC_ROOT
.
Then run the following command to evaluate the semantic segmentation results. The results will be saved in the results
folder, where the mean recall mrecall
is the mAcc and fmiou
is the F-mIoU reported in the paper.
# CoceptGraphs (without open-vocab detector)
python scripts/eval_replica_semseg.py \
--replica_root $REPLICA_ROOT \
--replica_semantic_root $REPLICA_SEMANTIC_ROOT \
--n_exclude 6 \
--pred_exp_name none_overlap_maskconf0.95_simsum1.2_dbscan.1_merge20_masksub
# On the ConceptGraphs-Detect (Grounding-DINO as the object detector)
python scripts/eval_replica_semseg.py \
--replica_root $REPLICA_ROOT \
--replica_semantic_root $REPLICA_SEMANTIC_ROOT \
--n_exclude 6 \
--pred_exp_name ram_withbg_allclasses_overlap_maskconf0.25_simsum1.2_dbscan.1_masksub
Ensure that the openai
package is installed and that your APIKEY is set. We recommend using GPT-4, since GPT-3.5 often produces inconsistent results on this task.
export OPENAI_API_KEY=<your GPT-4 API KEY here>
Also note that you may need to make the following change at this line in the original LLaVa repo to run the following commands.
# if output_ids[0, -keyword_id.shape[0]:] == keyword_id:
# return True
if torch.equal(output_ids[0, -keyword_id.shape[0]:], keyword_id):
return True
Then run the following commands sequentially to extract per-object captions and build the 3D scene graph.
SCENE_NAME=room0
PKL_FILENAME=output.pkl.gz # Change this to the actual output file name of the pkl.gz file
python scenegraph/build_scenegraph_cfslam.py \
--mode extract-node-captions \
--cachedir ${REPLICA_ROOT}/${SCENE_NAME}/sg_cache \
--mapfile ${REPLICA_ROOT}/${SCENE_NAME}/pcd_saves/${PKL_FILENAME}
python scenegraph/build_scenegraph_cfslam.py \
--mode refine-node-captions \
--cachedir ${REPLICA_ROOT}/${SCENE_NAME}/sg_cache \
--mapfile ${REPLICA_ROOT}/${SCENE_NAME}/pcd_saves/${PKL_FILENAME}
python scenegraph/build_scenegraph_cfslam.py \
--mode build-scenegraph \
--cachedir ${REPLICA_ROOT}/${SCENE_NAME}/sg_cache \
--mapfile ${REPLICA_ROOT}/${SCENE_NAME}/pcd_saves/${PKL_FILENAME}
Then the object map with scene graph can be visualized using the following command.
- Press
g
to show the scene graph. - Press "+" and "-" to increase and decrease the size of point cloud for better visualization.
python scripts/visualize_cfslam_results.py \
--result_path ${REPLICA_ROOT}/${SCENE_NAME}/sg_cache/map/scene_map_cfslam_pruned.pkl.gz \
--edge_file ${REPLICA_ROOT}/${SCENE_NAME}/sg_cache/cfslam_object_relations.json
During the development stage, we performed some experiments on the AI2Thor dataset. Upon request, now we provide the code and instructions for these experiments. However, note that we didn't perform any quantitative evaluation on AI2Thor. And because of domain gap, performance of ConceptGraphs may be worse than other datasets reported.
Use our own fork, where some changes were made to record the interaction trajectories.
cd .. # go back to the root folder CFSLAM
git clone [email protected]:georgegu1997/ai2thor.git
cd ai2thor
git checkout main5.0.0
pip install -e .
# This is for the ProcThor dataset.
pip install ai2thor-colab prior --upgrade
If you meet error saying Could not load the Qt platform plugin "xcb"
later on, it probably means that is some weird issue with opencv-python
and opencv-python-headless
. Try uninstalling them and install one of them back.
- Use
$AI2THOR_DATASET_ROOT
as the directory ai2thor dataset and save it to a variable. Also set the scene used from AI2Thor.
# Change this to run it in a different scene in AI2Thor environment
# train_3 is a scene from the ProcThor dataset, which containing multiple rooms in one house
SCENE_NAME=train_3
# The following scripts need to be run in the conceptgraph folder
cd ./conceptgraph
- Generate a densely captured grid map for the selected scene.
# Uniform sample camera locations (XY + Yaw)
python scripts/generate_ai2thor_dataset.py --dataset_root $AI2THOR_DATASET_ROOT --scene_name $SCENE_NAME --sample_method uniform --n_sample -1 --grid_size 0.5
# Uniform sample + randomize lighting
python scripts/generate_ai2thor_dataset.py --dataset_root $AI2THOR_DATASET_ROOT --scene_name $SCENE_NAME --sample_method uniform --n_sample -1 --grid_size 0.5 --save_suffix randlight --randomize_lighting
- Generate a human-controlled trajectory for the selected scene. (GUI and keyboard interaction needed)
# Interact generation and save trajectory files.
# This line will open up a Unity window. You can control the agent with arrow keys in the terminal window.
python scripts/generate_ai2thor_dataset.py --dataset_root $AI2THOR_DATASET_ROOT --scene_name $SCENE_NAME --interact
# Generate observations from the saved trajectory file
python scripts/generate_ai2thor_dataset.py --dataset_root $AI2THOR_DATASET_ROOT --scene_name $SCENE_NAME --sample_method from_file
- Generate a trajectory with object randomly moved.
MOVE_RATIO=0.25
RAND_SUFFIX=mv${MOVE_RATIO}
python scripts/generate_ai2thor_dataset.py \
--dataset_root $AI2THOR_DATASET_ROOT \
--scene_name $SCENE_NAME \
--interact \
--save_suffix $RAND_SUFFIX \
--randomize_move_moveable_ratio $MOVE_RATIO \
--randomize_move_pickupable_ratio $MOVE_RATIO