Shape-Guided icon indicating copy to clipboard operation
Shape-Guided copied to clipboard

Shape-Guided Dual-Memory Learning for 3D Anomaly Detection [ICML2023]

Results 11 Shape-Guided issues
Sort by recently updated
recently updated
newest added

![image](https://github.com/jayliu0313/Shape-Guided/assets/68214694/e4cb0030-e38f-4862-b346-62e145e24f3a) Why there is a rt in MVTec3DPreTrain?

Hey, @jayliu0313 Can we train the 3d model one sample a time but one patch a time? I think the pointnet and the NIF in Shape-Guided is designed for a...

![image](https://github.com/jayliu0313/Shape-Guided/assets/68214694/d097037a-b2de-4e0a-9216-5716aab12327) ![image](https://github.com/jayliu0313/Shape-Guided/assets/68214694/a65c9b95-8497-4fa3-a500-ef8f168b6dde) Hey, @jayliu0313 In the paper and code, I think the 'overlapping ratio' refers to the parameter 'group mul'. Have you conducted ablation experiments on the parameter of overlapping...

Why not train 3D model for each catergory respectively? I think if do so, the anomaly will be detected more easier.

``` def get_feature(self, points_all, points_idx, data_id, split='test'): total_feature = None total_rgb_feature_indices = [] for patch in range(len(points_all)): points = points_all[patch].reshape(-1, self.POINT_NUM, 3) indices = points_idx[patch].reshape(self.POINT_NUM) # compute the correspoding location...

``` def get_data_loader(split, class_name, img_size, datasets_path, grid_path, shuffle=False): if split in ['train', 'validation']: dataset = MVTec3DTrain(split=split, class_name=class_name, img_size=img_size, grid_path=grid_path) elif split in ['test']: dataset = MVTec3DTest(class_name=class_name, img_size=img_size, dataset_path=datasets_path, grid_path=grid_path) data_loader...

>To calibrate the distributions of scores, we align the mean ±3× standard deviation of RGB scores with the mean ±3× standard deviation of SDF scores by applying an affine transformation...

According to the paper, Pointnets (2017) are used for the 3D feature extraction. However, here a modified PCP model is used. Could you explain why it is so and why...

Thank you very much for your excellent work! I recently collected a lot of point cloud data in xyz format, which is different from RGB-Depth format. My requirement is object-level...

(base) anywhere3090l@3090l:~/Desktop/Shape-Guided$ python tools/preprocessing.py /home/anywhere3090l/Desktop/Shape-Guided/mvtec3d Traceback (most recent call last): File "/home/anywhere3090l/Desktop/Shape-Guided/tools/preprocessing.py", line 8, in import utils.mvtec3d_util as mvt_util ModuleNotFoundError: No module named 'utils.mvtec3d_util'