3D Shape Completion with Multi-View Consistent Inference
Authors: Tao Hu, Zhizhong Han, Matthias Zwicker10997-11004
AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results demonstrate that our method completes shapes more accurately than previous techniques. |
| Researcher Affiliation | Academia | Tao Hu, Zhizhong Han, Matthias Zwicker Department of Computer Science University of Maryland, College Park {taohu, zwicker}@cs.umd.edu, h312h@umd.edu |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any statement or link regarding the public availability of its source code. |
| Open Datasets | Yes | We take 3D models from Shape Net (Chang et al. 2015). |
| Dataset Splits | No | The paper mentions using a 'training dataset' and a 'test dataset' from previous works but does not explicitly provide details about training, validation, and test splits (e.g., percentages, counts, or explicit validation set mention). |
| Hardware Specification | No | The paper does not provide specific details about the hardware (e.g., GPU/CPU models, memory) used for running its experiments. |
| Software Dependencies | No | The paper mentions various techniques and frameworks like U-Net, GAN, and batch normalization, but does not provide specific version numbers for any software dependencies or libraries. |
| Experiment Setup | Yes | We optimize the shape descriptor z for 100 gradient descent steps, and we take z with the smallest consistency loss in the last 10 steps as z . It should be mentioned that since the gradients of z are small, we use a large learning rate of 0.2. ... According to the comparison, we select L2 distance to calculate generator loss, and set μ = 1 in the following experiments. ... The average CD is lower when we increase the size of the depth-buffer, as shown in Table 2 (a)... Given a depth-buffer size of 5 × 5, Fig. 6 (e) to (g) show that the consistency loss increases when J = 3 or J = 5... |