World-GAN: a Generative Model for Minecraft Worlds

Awiszus, Maren; Schubert, Frederik; Rosenhahn, Bodo

doi:10.1109/cog52621.2021.9619133

Cited by 14 publications

(5 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Each token contains a type of object or part of it, such as grass, wood, and air, represented by a semantic label. (Awiszus, Schubert, and Rosenhahn 2021) proposed World-GAN, an architecture similar to SinGAN that learns from a single level in an unsupervised way. (Sudhakaran et al 2021) adopted Neural Cellular Automata (NCA) (Mordvintsev et al 2020) in reconstructing Minecraft artefacts.…”

Section: Procedural Content Generationmentioning

confidence: 99%

See 1 more Smart Citation

Procedural Level Generation with Diffusion Models from a Single Example

Dai,

Zhu,

et al. 2024

AAAI

View full text Add to dashboard Cite

Level generation is a central focus of Procedural Content Generation (PCG), yet deep learning-based approaches are limited by scarce training data, i.e., human-designed levels. Despite being a dominant framework, Generative Adversarial Networks (GANs) exhibit a substantial quality gap between generated and human-authored levels, alongside rising training costs, particularly with increasing token complexity. In this paper, we introduce a diffusion-based generative model that learns from just one example. Our approach involves two core components: 1) an efficient yet expressive level representation, and 2) a latent denoising network with constrained receptive fields. To start with, our method utilizes token semantic labels, similar to word embeddings, to provide dense representations. This strategy not only surpasses one-hot encoding in representing larger game levels but also improves stability and accelerates convergence in latent diffusion. In addition, we adapt the denoising network architecture to confine the receptive field to localized patches of the data, aiming to facilitate single-example learning. Extensive experiments demonstrate that our model is capable of generating stylistically congruent samples of arbitrary sizes compared to manually designed levels. It suits a wide range of level structures with fewer artifacts than GAN-based approaches. The source code is available at https://github.com/shiqi-dai/diffusioncraft.

show abstract

Section: Procedural Content Generationmentioning

confidence: 99%

“…Likewise, this implies that there is no need to derive token representations from a Language Model, since while their embeddings are semantically rich, they are overly complex. A dimension of d = 32 suffices for our task, as evidenced by WorldGAN (Awiszus, Schubert, and Rosenhahn 2021).…”

Section: Level Representationmentioning

confidence: 99%

Procedural Level Generation with Diffusion Models from a Single Example

Dai,

Zhu,

et al. 2024

AAAI

View full text Add to dashboard Cite

show abstract

“…Neural cellular automata have been used to create structures such caves, buildings, and trees with increasing complexity and ability to regenerate and repair themselves [38]. Generation of the world itself has also been researched through works such as World-GAN [39] which attempts to address the shortcomings of the static world generator bundled with Minecraft through generative adversarial networks. We position this work to fill the gap in the research space for interesting and creative buildings for Minecraft settlements.…”

Section: Minecraft Settlement Generationmentioning

confidence: 99%

Open-Ended Evolution for Minecraft Building Generation

2023

View full text Add to dashboard Cite

This paper proposes a procedural content generator which evolves Minecraft buildings according to an open-ended and intrinsic definition of novelty. To realize this goal we evaluate individuals' novelty in the latent space using a 3D autoencoder, and alternate between phases of exploration and transformation. During exploration the system evolves multiple populations of CPPNs through CPPN-NEAT and constrained novelty search in the latent space (defined by the current autoencoder). We apply a set of repair and constraint functions to ensure candidates adhere to basic structural rules and constraints during evolution. During transformation, we reshape the boundaries of the latent space to identify new interesting areas of the solution space by retraining the autoencoder with novel content. In this study we evaluate five different approaches for training the autoencoder during transformation and its impact on populations' quality and diversity during evolution. Our results show that by retraining the autoencoder we can achieve better open-ended complexity compared to a static model, which is further improved when retraining using larger datasets of individuals with diverse complexities.

show abstract

“…In the wake of strikingly realistic randomly generated 2D images [ 1 , 2 ], there is a mounting expectation for generative models to replicate the same success when automatically synthesizing 3D objects. Demand for such models arises in various domains, ranging from rapid design and prototyping for the manufacturing sector to the entertainment industry [ 3 , 4 , 5 , 6 , 7 ]. Hence, the research focus is shifting towards 3D deep generative models, as rich and flexible 3D representations are rapidly emerging.…”

Section: Introductionmentioning

confidence: 99%

“…Many of these applications benefit from or require quick generation of the desired 3D object. One example is a machine learning model that fabricates levels in a video game to grant the user a unique experience [ 5 , 6 ]. Real-time computation allows the random generation of items or characters during the play.…”

Section: Introductionmentioning

confidence: 99%

PCGen: A Fully Parallelizable Point Cloud Generative Model

Vercheval,

Royen,

Munteanu

et al. 2024

Sensors

View full text Add to dashboard Cite

Generative models have the potential to revolutionize 3D extended reality. A primary obstacle is that augmented and virtual reality need real-time computing. Current state-of-the-art point cloud random generation methods are not fast enough for these applications. We introduce a vector-quantized variational autoencoder model (VQVAE) that can synthesize high-quality point clouds in milliseconds. Unlike previous work in VQVAEs, our model offers a compact sample representation suitable for conditional generation and data exploration with potential applications in rapid prototyping. We achieve this result by combining architectural improvements with an innovative approach for probabilistic random generation. First, we rethink current parallel point cloud autoencoder structures, and we propose several solutions to improve robustness, efficiency and reconstruction quality. Notable contributions in the decoder architecture include an innovative computation layer to process the shape semantic information, an attention mechanism that helps the model focus on different areas and a filter to cover possible sampling errors. Secondly, we introduce a parallel sampling strategy for VQVAE models consisting of a double encoding system, where a variational autoencoder learns how to generate the complex discrete distribution of the VQVAE, not only allowing quick inference but also describing the shape with a few global variables. We compare the proposed decoder and our VQVAE model with established and concurrent work, and we prove, one by one, the validity of the single contributions.

show abstract

World-GAN: a Generative Model for Minecraft Worlds

Cited by 14 publications

References 22 publications

Procedural Level Generation with Diffusion Models from a Single Example

Procedural Level Generation with Diffusion Models from a Single Example

Open-Ended Evolution for Minecraft Building Generation

PCGen: A Fully Parallelizable Point Cloud Generative Model

Contact Info

Product

Resources

About