Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
Shape-Guided Dual-Memory Learning for 3D Anomaly Detection
Authors: Yu-Min Chu, Chieh Liu, Ting-I Hsieh, Hwann-Tzong Chen, Tyng-Luh Liu
ICML 2023 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate our method on MVTec 3DAD (Bergmann & Sattlegger, 2022), which provides ten different categories of 3D objects for 2D+3D anomaly detection. |
| Researcher Affiliation | Collaboration | 1Department of Computer Science, National Tsing Hua University, Hsinchu, Taiwan 2Aeolus Robotics, Taipei, Taiwan 3Institute of Information Science, Academia Sinica, Taipei, Taiwan. |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not include an unambiguous statement about releasing source code for the described methodology or a direct link to a code repository. |
| Open Datasets | Yes | We evaluate our method on MVTec 3DAD (Bergmann & Sattlegger, 2022), which provides ten different categories of 3D objects for 2D+3D anomaly detection. |
| Dataset Splits | Yes | MVTec 3D-AD contains 2,656 training, 294 validation, and 1,197 test samples. |
| Hardware Specification | Yes | Table 3. Comparison of inference time (second) per sample, frame per second (FPS), the number of features (No F), and the percentage of RGB memory usage on an Nvidia GTX 1080 GPU. |
| Software Dependencies | No | The paper mentions software components like "Point Net", "NIF", "Image Net pretrained Res Net", "Wide Res Net-50-2", and "Patch Core" but does not provide specific version numbers for any of them. |
| Experiment Setup | Yes | We set the learning rate and batch size to 0.0001 and 32, respectively, which empirically achieves efficient convergence. |