Visual Instruction Generation on ShowHowTo (test)
70Step FaithfulnessStable Diffusion
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Stable Diffusionsetup=zero-shot, conditioned_on_input_image=false2024.12 | 70 | 3 | 44 | — | |
| ShowHowTosetup=zero-shot2024.12 | 52 | 34 | 42 | — | |
| Source sequences2024.12 | 50 | 100 | 56 | — | |
| GenHowTosetup=zero-shot, generation_strategy=iterative2024.12 | 49 | 13 | 27 | — | |
| StackedDiffusionsetup=zero-shot, generation_strategy=input masking2024.12 | 43 | 2 | 42 | — | |
| Phung et al.setup=zero-shot, backbone=Stable Diffusion [48], generation_strategy=input masking2024.12 | 36 | 3 | 38 | — | |
| InstructPix2Pixsetup=zero-shot, generation_strategy=iterative2024.12 | 25 | 17 | 25 | — | |
| AURORAsetup=zero-shot, generation_strategy=iterative2024.12 | 25 | 33 | 24 | — | |
| Random2024.12 | 19 | 2 | 1 | — | |
| Copy of the input image2024.12 | 19 | 62 | 39 | — | |
| AURORA2024.12 | — | — | — | 23.2 | |
| GenHowTo2024.12 | — | — | — | 28.3 | |
| InstructPix2Pix2024.12 | — | — | — | 37.8 | |
| Phung et al.2024.12 | — | — | — | 27.8 | |
| ShowHowTo2024.12 | — | — | — | 12.4 | |
| StackedDiffusion2024.12 | — | — | — | 34.6 |