domaingen
domaingen copied to clipboard
vidit09
CLIP the Gap CVPR 2023
I have tried to replicate the results using the official weights for training, but my results differ from yours. I believe that in the second stage, Aj should be frozen,...
I trained and got the final trained model, but I got the problem when want to apply the model for image testing. Could you give me the detail code for...
**hello，This is the code I used during testing. May I ask if the configuration file is correct, especially the name of the test dataset.** --------------------------------------------------------------------------------------------------------- BASE_YAML: "COCO-Detection/faster_rcnn_R_101_C4_3x.yaml" DATASETS: TRAIN: ("daytime_clear_train",)...
I use daytime_clear_train as the training set, Day-Clear_test，Night-Clear_train，Dusk-Rainy_train，Night-Rainy_train，Day-Foggy_train As a test set, but compared to the original paper, my experimental results showed significant drops in all datasets except for the...
Where is semantic augmentation by difference in text prompts implemented?
Could you please share how you applied the CLIP initialization to the ori RN101? Could you also let me know where I can find the corresponding code?
In section 4.2 Implementation Details :"We initialize the detector with CLIP pre-trained weights" I can't find the code position
How to get the evaluation results of other target domains such as night_rainy? Should I directly change the configuration file or need to register the data set? Are other data...
Hello! Thanks for open-sourcing the code! I'm having trouble reproducing your results on the night rainy task.Specifically, in your paper, it shows that on the night rainy task, the AP...
