Adds data collection pipeline (#40)

# Description

Adds data collection pipeline. 

Fixes #7 #14 #36 

## Type of change

- New feature (non-breaking change which adds functionality)

## Checklist

- [x] I have run the [`pre-commit` checks](https://pre-commit.com/) with
`./formatter.sh`
- [x] I have made corresponding changes to the documentation
- [x] My changes generate no new warnings

README.md

@@ -92,10 +92,10 @@ Here an overview of the steps involved in training the policy.
 For more detailed instructions, please refer to [TRAINING.md](TRAINING.md).
 
 0. Training Data Generation <br>
-Training data is generated from the [Matterport 3D](https://github.com/niessner/Matterport), [Carla](https://carla.org/) and [NVIDIA Warehouse](https://docs.omniverse.nvidia.com/isaacsim/latest/tutorial_static_assets.html) using developed Isaac Sim Extension, the extensions are part of a new internal project (``isaac-nav-suite``) and will be open sourced with that project. In the case that you require an earlier access, please contact us via mail.
+Training data is generated from the [Matterport 3D](https://github.com/niessner/Matterport), [Carla](https://carla.org/) and [NVIDIA Warehouse](https://docs.omniverse.nvidia.com/isaacsim/latest/tutorial_static_assets.html) using IsaacLab. For detailed instruction on how to install the extension and run the data collection script, please see [here](omniverse/README.md)
 
 1. Build Cost-Map <br>
-The first step in training the policy is to build a cost-map from the available depth and semantic data. A cost-map is a representation of the environment where each cell is assigned a cost value indicating its traversability. The cost-map guides the optimization, therefore, is required to be differentiable. Cost-maps are built using the [cost-builder](viplanner/cost_builder.py) with configs [here](viplanner/config/costmap_cfg.py), given a pointcloud of the environment with semantic information (either from simultion or real-world information).
+The first step in training the policy is to build a cost-map from the available depth and semantic data. A cost-map is a representation of the environment where each cell is assigned a cost value indicating its traversability. The cost-map guides the optimization, therefore, is required to be differentiable. Cost-maps are built using the [cost-builder](viplanner/cost_builder.py) with configs [here](viplanner/config/costmap_cfg.py), given a pointcloud of the environment with semantic information (either from simultion or real-world information). The point-cloud of the simulated environments can be generated with the [reconstruction-script](viplanner/depth_reconstruct.py) with config [here](viplanner/config/costmap_cfg.py).
 
 2. Training <br>
 Once the cost-map is constructed, the next step is to train the policy. The policy is a machine learning model that learns to make decisions based on the depth and semantic measurements. An example training script can be found [here](viplanner/train.py) with configs [here](viplanner/config/learning_cfg.py)

TRAINING.md

@@ -2,6 +2,12 @@
 
 Here an overview of the steps involved in training the policy is provided.
 
+
+## Data Generation
+
+For the data generation, please follow the instruction given in [here](omniverse/README.md).
+
+
 ## Cost-Map Building
 
 Cost-Map building is an essential step in guiding optimization and representing the environment.
@@ -28,13 +34,14 @@ If depth and semantic images of the simulation are available, then first 3D reco
         ├── xxxx.png                            # images saved with 4 digits, e.g. 0000.png
     ```
 
-    when both depth and semantic images are available, then define sem_suffic and depth_suffix in ReconstructionCfg to differentiate between the two with the following structure:
+    In the case that the semantic and depth images have an offset in their position (as typical on some robotic platforms),
+    define a `sem_suffic` and `depth_suffix` in `ReconstructionCfg` to differentiate between the two with the following structure:
 
     ``` graphql
     env_name
     ├── camera_extrinsic{depth_suffix}.txt      # format: x y z qx qy qz qw
     ├── camera_extrinsic{sem_suffix}.txt        # format: x y z qx qy qz qw
-    ├── intrinsics.txt                          # P-Matrix for intrinsics of depth and semantic images
+    ├── intrinsics.txt                          # P-Matrix for intrinsics of depth and semantic images (depth first)
     ├── depth                                   # either png and/ or npy, if both npy is used
     |   ├── xxxx{depth_suffix}.png              # images saved with 4 digits, e.g. 0000.png
     |   ├── xxxx{depth_suffix}.npy              # arrays saved with 4 digits, e.g. 0000.npy
@@ -49,7 +56,7 @@ If depth and semantic images of the simulation are available, then first 3D reco
 
 3. **Cost-Building** <br>
 
-    Fully automated, either a geometric or semantic cost map can be generated running the following command:
+    Either a geometric or semantic cost map can be generated running the following command:
 
     ```
     python viplanner/cost_builder.py
@@ -72,7 +79,6 @@ If depth and semantic images of the simulation are available, then first 3D reco
     ```
 
 
-
 ## Training
 
 Configurations of the training given in [TrainCfg](viplanner/config/learning_cfg.py). Training can be started using the example training script [train.py](viplanner/train.py).

omniverse/README.md

@@ -12,7 +12,7 @@
 [![pre-commit](https://img.shields.io/badge/pre--commit-enabled-brightgreen?logo=pre-commit&logoColor=white)](https://pre-commit.com/)
 [![License](https://img.shields.io/badge/license-BSD--3-yellow.svg)](https://opensource.org/licenses/BSD-3-Clause)
 
-The ViPlanner Omniverse Extension offers a testing environment for ViPlanner.
+The ViPlanner Omniverse Extension offers a testing environment for ViPlanner and includes the data collection pipeline.
 Within NVIDIA Isaac Sim as a photorealistic simulator and using [IsaacLab](https://isaac-sim.github.io/IsaacLab/), this extension provides an assessment tool for ViPlanner's performance across diverse environments.
 
 
@@ -62,16 +62,9 @@ It is necessary to comply with PEP660 for the install. This requires the followi
   ./isaaclab.sh -p -m pip install --upgrade setuptools
   ```
 
-## Usage
-
-A demo script is provided to run the planner in three different environments: [Matterport](https://niessner.github.io/Matterport/), [Carla](https://carla.org//), and [NVIDIA Warehouse](https://docs.omniverse.nvidia.com/isaacsim/latest/features/environment_setup/assets/usd_assets_environments.html#warehouse).
-In each scenario, the goal is represented as a movable cube within the environment.
-
-To run the demo, download the model: [[checkpoint](https://drive.google.com/file/d/1PY7XBkyIGESjdh1cMSiJgwwaIT0WaxIc/view?usp=sharing)] [[config](https://drive.google.com/file/d/1r1yhNQAJnjpn9-xpAQWGaQedwma5zokr/view?usp=sharing)] and the environment files. Then adjust the paths (marked as `${USER_PATH_TO_USD}`) in the corresponding config files.
+## Download the Simulation Environments
 
 ### Matterport
-[Config](./extension/omni.viplanner/omni/viplanner/config/matterport_cfg.py)
-
 To download Matterport datasets, please refer to the [Matterport3D](https://niessner.github.io/Matterport/) website. The dataset should be converted to USD format using Isaac Sim by executing the following steps:
 
 1. Run the `convert_mesh.py` script to convert the `.obj` file (located under `matterport_mesh`) to `USD`. With the recent update of the asset converter script, use the resulting `*_non_metric.usd` file.
@@ -92,6 +85,20 @@ To download Matterport datasets, please refer to the [Matterport3D](https://nies
 top left corner, select `Show by Type -> Physics -> Colliders` and set the value to `All` ). The colliders should be visible as pink linkes. In the case that no colliders are presented, select the mesh in the stage,
 go the `Property` section and click `Add -> Physics -> Colliders Preset`. Then save the asset.
 
+### Carla
+We provide an already converted asset of the `Town01` of Carla. It can be downloaded as USD asset: [Download USD Link](https://drive.google.com/file/d/1wZVKf2W0bSmP1Wm2w1XgftzSBx0UR1RK/view?usp=sharing)
+
+
+## Planer Demo
+
+A demo script is provided to run the planner in three different environments: [Matterport](https://niessner.github.io/Matterport/), [Carla](https://carla.org//), and [NVIDIA Warehouse](https://docs.omniverse.nvidia.com/isaacsim/latest/features/environment_setup/assets/usd_assets_environments.html#warehouse).
+In each scenario, the goal is represented as a movable cube within the environment.
+
+To run the demo, download the model: [[checkpoint](https://drive.google.com/file/d/1PY7XBkyIGESjdh1cMSiJgwwaIT0WaxIc/view?usp=sharing)] [[config](https://drive.google.com/file/d/1r1yhNQAJnjpn9-xpAQWGaQedwma5zokr/view?usp=sharing)] and the environment files. Then adjust the paths (marked as `${USER_PATH_TO_USD}`) in the corresponding config files.
+
+### Matterport
+[Config](./extension/omni.viplanner/omni/viplanner/config/matterport_cfg.py)
+
 The demo uses the **2n8kARJN3HM** scene from the Matterport dataset. A preview is available [here](https://aspis.cmpt.sfu.ca/scene-toolkit/scans/matterport3d/houses).
 
 ```
@@ -100,7 +107,7 @@ cd IsaacLab
 ```
 
 ### Carla
-[Download USD Link](https://drive.google.com/file/d/1wZVKf2W0bSmP1Wm2w1XgftzSBx0UR1RK/view?usp=sharing) | [Config](./extension/omni.viplanner/omni/viplanner/config/carla_cfg.py)
+[Config](./extension/omni.viplanner/omni/viplanner/config/carla_cfg.py)
 
 ```
 cd IsaacLab
@@ -115,7 +122,14 @@ cd IsaacLab
 ./isaaclab.sh -p <path-to-viplanner-repo>/omniverse/standalone/viplanner_demo.py --scene warehouse --model_dir <path-to-model-download-dir>
 ```
 
-## Data Collection and Evaluation
+## Data Collection
+
+The training data is generated from different simulation environments. After they have been downloaded and converted to USD, adjust the paths (marked as `${USER_PATH_TO_USD}`) in the corresponding config files ([Carla](./extension/omni.viplanner/omni/viplanner/config/carla_cfg.py) and [Matterport](./extension/omni.viplanner/omni/viplanner/config/matterport_cfg.py)).
+The rendered viewpoints are collected by executing
+
+```
+cd IsaacLab
+./isaaclab.sh -p <path-to-viplanner-repo>/omniverse/standalone/viplanner_demo.py --scene <matterport/carla/warehouse> --num_samples <how-many-viewpoints>
+```
 
-The data collection is currently included in a new internal project and will be released with this project in the future.
-If you require the code, please contact us per mail.
+To test that the data has been correctly extracted, please run the 3D reconstruction and see if the results fits to the simulated environment.

omniverse/extension/omni.isaac.matterport/omni/isaac/matterport/domains/matterport_raycast_camera.py

@@ -62,6 +62,7 @@ def _initialize_impl(self):
         # More Information: https://github.com/niessner/Matterport/blob/master/data_organization.md#house_segmentations
         mapping = pd.read_csv(DATA_DIR + "/mappings/category_mapping.tsv", sep="\t")
         self.mapping_mpcat40 = torch.tensor(mapping["mpcat40index"].to_numpy(), device=self._device, dtype=torch.long)
+        self.classes_mpcat40 = pd.read_csv(DATA_DIR + "/mappings/mpcat40.tsv", sep="\t")["mpcat40"].to_numpy()
         self._color_mapping()
 
     def _color_mapping(self):

omniverse/extension/omni.viplanner/data/warehouse/keyword_mapping.yml

@@ -5,6 +5,7 @@
 # SPDX-License-Identifier: BSD-3-Clause
 
 floor:
+  - SM_Floor0
   - SM_Floor1
   - SM_Floor2
   - SM_Floor3
@@ -31,6 +32,7 @@ ceiling:
 
 static:
   - LampCeiling
+  - Section
   - SM_FloorDecal
   - SM_FireExtinguisher
 

omniverse/extension/omni.viplanner/omni/viplanner/collectors/__init__.py

@@ -0,0 +1,8 @@
+# Copyright (c) 2023-2024, ETH Zurich (Robotics Systems Lab)
+# Author: Pascal Roth
+# All rights reserved.
+#
+# SPDX-License-Identifier: BSD-3-Clause
+
+from .viewpoint_sampling import ViewpointSampling
+from .viewpoint_sampling_cfg import ViewpointSamplingCfg