ZeroShape: The Inference on AI-Generated Images

Table of Links

Abstract and 1 Introduction

2. Related Work

3. Method and 3.1. Architecture

3.2. Loss and 3.3. Implementation Details

4. Data Curation

4.1. Training Dataset

4.2. Evaluation Benchmark

5. Experiments and 5.1. Metrics

5.2. Baselines

5.3. Comparison to SOTA Methods

5.4. Qualitative Results and 5.5. Ablation Study

6. Limitations and Discussion

7. Conclusion and References

A. Additional Qualitative Comparison

B. Inference on AI-generated Images

C. Data Curation Details

B. Inference on AI-generated Images

We present additional results of ZeroShape using images generated with DALL·E 3. To test the out-of-domain generalization ability, we generate images of imaginary objects as the input to our model (see Fig. 10). Despite the domain gap to realistic or rendered images, ZeroShape can faithfully recover the global shape structure and accurately follow the local geometry cues from the input image. These results also demonstrate the potential of using ZeroShape in a text-based 3D generation workflow.