Closed ashawkey closed 2 years ago
Failure cases:
I tried DSLR photos of a panda, got a similar failed result. Curious: any guess why it failed?
Maybe it's because small prey animals have adapted shape and color patterns to make it difficult to guess their true orientation and so which direction they'll scurry or hop away.
This is my first tests, each take about 90 minutes:
--text "a hamburger"
https://user-images.githubusercontent.com/65137/197421283-d31aa5ee-efeb-45fb-9dd5-b16e46be74a1.mp4
--text "jeweled dagger"
https://user-images.githubusercontent.com/65137/197421294-a7950c2c-e7a0-40a4-8307-0eb1fa0fc78f.mp4
Certainly, a lot more to tune and play with.
Some results from the pure-pytorch vanilla NeRF backbone:
@ashawkey Hi, can you please provide these commands to reproduce the new results video here?
https://github.com/ashawkey/stable-dreamfusion/tree/8d7899ae758e0f683a1cef51b876f2a37cba0c55
@elliottzheng Hi, they are run with the default configs, with prompts like "a superman". You could also refer to this: https://github.com/ashawkey/stable-dreamfusion/issues/128#issuecomment-1413593621.
https://user-images.githubusercontent.com/25863658/197331690-2c9f7f55-e571-4e66-90f3-fd3580c9a88f.mp4
https://user-images.githubusercontent.com/25863658/197331713-ba7d340e-b23e-47ba-90ef-54fa682edda5.mp4
https://user-images.githubusercontent.com/25863658/197331760-59a02698-f5be-4867-a303-d6a7ccc9a0d7.mp4
https://user-images.githubusercontent.com/25863658/197331771-7bb4cdb1-ed17-428e-934c-5c694d0fcd8a.mp4
https://user-images.githubusercontent.com/25863658/194847191-77740afc-9a51-4646-9c1f-7f59cdc7b13d.mp4
https://user-images.githubusercontent.com/25863658/194846953-ce5e343e-58d1-4c97-8af7-c3d63bf1d894.mp4
https://user-images.githubusercontent.com/25863658/194827909-b20269da-9e9b-477e-9926-06925d3dfea7.mp4
https://user-images.githubusercontent.com/25863658/194284983-623a4ade-0029-490e-b038-c7dfa83bb405.mp4
https://user-images.githubusercontent.com/25863658/194285798-711feaf9-fc5a-4098-b7b5-bf2d2b0758b7.mp4
https://user-images.githubusercontent.com/25863658/194288254-facfa486-73c4-43c3-819b-0739ca950086.mp4