Ask HN: Has anyone trained GAN on larger resolution images?
Has anyone been able to train GAN on larger resolution images like 640x480
Has anyone been able to train GAN on larger resolution images like 640x480
At pikup.ai, we are building the new age autonomous store, using Artificial Intelligence, Computer Vision and Sensor Fusion. We believe that after vending machines, malls, e-commerce, this will be the biggest disruption in the world of retail. We are a team of world class tech and business hackers with renowned and significant accomplishments who are looking for like minded people to join us to...
Let's say you you two object detectors -FasterRCNN with Resnet backbone and SSD with Resnet and focal loss.Both have similar loss but difference in mAP. How do you compare these models. Does low loss imply both understand the dataset equally good but the variance aka generalisation is better in the one with high mAp
What are good data augmentation techniques which can work with 3d data specifically depth being third dimension. Also, any pointers on tools around them would be of great help.
Any existing tools to label a new dataset with person keypoints.
OPening this as a discussion thread to understand how Amazon GO AI works.
Hackers,i am looking for some reference to understand how to apply transfer learning for detecting certain human activities in a live video feed.