-
*Initial cost estimate*: 2 days
*Initial contacts*: @trylek, @jkoritzinsky, @davidwrighton
OS: **Windows**
Architecture: **x86**
Configuration: **Release**
Repro steps (non-deterministic): **b…
-
This is on an Ubuntu 11.10 (Oneiric) 32-bit system. Here's the kernel config:
https://gist.github.com/3058026
```
$ uname -a
Linux oneiric32 3.0.0-12-generic #20-Ubuntu SMP Fri Oct 7 14:50:42 UTC 20…
-
Hi xuanqing, I just read you paper "Towards Robust Neural Networks via Random Self-ensemble" and I do think RSE is a brilliant idea.
I have one simple question about your experiement setting. You men…
-
In Paper, in the equation 4, what is the difference between d and z ? z is the depth in camera coordinate, and what is the phisical meaning of d ?
In equation 3, you said that bx,by and bz denote th…
-
Hello author, may I ask if I have downloaded the APRICOT MASK dataset and would like to test the SAC method on it? By using the PGD method to generate digital adversarial samples on the coco dataset, …
-
Hi, in the log you provide, the loaded pretrained model is called pgd.pth while the provided model is fcos3d.pth, are these two models the same?
-
Despite closing and reopening the PolyGot program, I cannot remove Definiteness from the declension types of Verb.
MacOS Ventura 13.2.1
[Kauvi'ali_before pruning_202307271043.pgd.zip](https://gi…
-
I encountered an error while trying to run the ART_attack.py script for my YOLOv5 object detection model. The error message reads:
The code:
```python
model = torch.hub.load('ultralytics/yolov5', '…
-
I downloaded the model weight you released in [Google drive folder](https://drive.google.com/open?id=1W2zGHyxTPgHhWln1kpHK5h-HY9kwfKfl) and found that it is not robust.
```
CUDA_VISIBLE_DEVICES=0,1…
-
Thank you for your amazing paper.
I have read your paper. You said that PGD keep top-K regions with highest quality score. As my understanding, for each ground-truth box, we will keep K **feature pix…