Add LevDoom and COOM benchmarks to the tools/projects list (#585)

* Add LevDoom and COOM benchmarks to the tools/projects list

* Added LevDoom and COOM to the third-party environments documentation page

* Modify the docs about third-party environments

---------

Co-authored-by: Marek Wydmuch <[email protected]>

README.md

@@ -156,6 +156,8 @@ Useful articles (for advanced users who want to create custom environments/scena
 - [Sample Factory](https://github.com/alex-petrenko/sample-factory) - A high-performance reinforcement learning framework for ViZDoom.
 - [EnvPool](https://github.com/sail-sg/envpool/) - A high-performance vectorized environment for ViZDoom.
 - [Obsidian](https://github.com/dashodanger/Obsidian) - Doom random map generator, a continuation of OBLIGE.
+- [LevDoom](https://github.com/TTomilin/LevDoom) - Generalization benchmark in ViZDoom featuring difficulty levels in visual complexity.
+- [COOM](https://github.com/hyintell/COOM) - Continual learning benchmark in ViZDoom offering task sequences with diverse objectives.
 
 
 ## Contributions

docs/environments/thirdParty.md

@@ -0,0 +1,24 @@
+# Third-party environments
+
+Here, we feature a selection of third-party libraries that build upon or complement ViZDoom,
+offering diverse environments and tools for reinforcement learning research and development.
+
+*Please note that the page contains environments that are not maintained by the ViZDoom Team or Farama Foundation.*
+
+*If you have a library that you would like to see featured here, please open a pull request or an issue on the [GitHub repository](https://github.com/Farama-Foundation/ViZDoom)*
+
+
+## LevDoom
+
+[LevDoom](https://github.com/TTomilin/LevDoom) is a benchmark for generalization in pixel-based deep reinforcement learning, offering environments with difficulty levels based on visual and gameplay modifications. It consists of 4 scenarios, each with 5 difficulty levels, that modify different aspects of the base environments, such as textures, obstacles, types, sizes, and rendering of different in-game entities, etc.
+
+LevDoom provides environments using Gymnasium API and is available through PyPi. For more details, please refer to the [CoG2022 paper](https://ieee-cog.org/2022/assets/papers/paper_30.pdf) and [GitHub repository](https://github.com/TTomilin/LevDoom).
+
+
+## COOM
+
+[COOM](https://github.com/hyintell/COOM) is a Continual Learning benchmark for embodied pixel-based RL, consisting of task sequences in visually distinct 3D environments with diverse objectives and egocentric perception. COOM is designed for task-incremental learning, in which task boundaries are clearly defined. It contains 8 scenarios, every with at least 2 difficulty levels that are combined into sequences of tasks for Continual Learning. The sequence length varies between 4, 8, and 16. COOM provides two types of task sequences:
+- Cross-domain sequences compose modified versions of the same scenario (e.g., changing textures, enemy types, view height, and adding obstacles) while keeping the objective consistent.
+- Cross-objective sequences contrast with Cross-Domain by changing both the visual characteristics and the objective for each task, requiring a more general policy from the agent for adequate performance.
+
+COOM provides environments using Gymnasium API and is available through PyPi. For more details, please refer to the [GitHub repository](https://github.com/hyintell/COOM).

docs/index.md

@@ -78,6 +78,7 @@ api/configurationFiles
 :caption: Environments
 
 environments/default
+environments/thirdParty
 environments/creatingCustom
 ```