Shap-E：生成有條件的3D隱式函數

摘要

我們提出了 Shap-E，一種用於3D資產的有條件生成模型。與最近關於3D生成模型的工作不同，這些模型產生單一輸出表示，Shap-E直接生成可以呈現為紋理網格和神經輻射場的隱式函數的參數。我們通過兩個階段來訓練Shap-E：首先，我們訓練一個編碼器，將3D資產確定性地映射到隱式函數的參數；其次，我們對編碼器的輸出訓練一個有條件擴散模型。當在大量配對的3D和文本數據集上進行訓練時，我們的模型能夠在幾秒內生成複雜且多樣的3D資產。與Point-E相比，Point-E是一種基於點雲的顯式生成模型，Shap-E收斂速度更快，盡管對建模更高維度、多表示輸出空間，但達到了相當或更好的樣本質量。我們在https://github.com/openai/shap-e上釋放模型權重、推理代碼和樣本。

English

We present Shap-E, a conditional generative model for 3D assets. Unlike recent work on 3D generative models which produce a single output representation, Shap-E directly generates the parameters of implicit functions that can be rendered as both textured meshes and neural radiance fields. We train Shap-E in two stages: first, we train an encoder that deterministically maps 3D assets into the parameters of an implicit function; second, we train a conditional diffusion model on outputs of the encoder. When trained on a large dataset of paired 3D and text data, our resulting models are capable of generating complex and diverse 3D assets in a matter of seconds. When compared to Point-E, an explicit generative model over point clouds, Shap-E converges faster and reaches comparable or better sample quality despite modeling a higher-dimensional, multi-representation output space. We release model weights, inference code, and samples at https://github.com/openai/shap-e.

Shap-E：生成有條件的3D隱式函數

Shap-E: Generating Conditional 3D Implicit Functions

摘要

Support