Haoxiang-Wang
commited on
Commit
•
dd3e2a5
1
Parent(s):
98f1da4
Update README.md
Browse files
README.md
CHANGED
@@ -280,7 +280,7 @@ Model Type:
|
|
280 |
Intended Users: | Generative AI developers for image and video generation models
|
281 |
Output: | Images/Videos and Latent Tokens
|
282 |
Describe how the model works: | Compresses and decompresses visual input (image/video).
|
283 |
-
Technical Limitations: |
|
284 |
Verified to have met prescribed NVIDIA quality standards: | Yes
|
285 |
Performance Metrics: | Peak Signal-to-Noise Ratio (PSNR), Structural Similarity (SSIM), Reconstruction Fréchet Video Distance (rFVD), Reconstruction Fréchet Inception Distance (rFID), Latency
|
286 |
Potential Known Risks: | Tokenizer's output can parse all forms of input, including what may be considered toxic, offensive, or indecent.
|
|
|
280 |
Intended Users: | Generative AI developers for image and video generation models
|
281 |
Output: | Images/Videos and Latent Tokens
|
282 |
Describe how the model works: | Compresses and decompresses visual input (image/video).
|
283 |
+
Technical Limitations: | Due to tokenizer compression limitations, some visual information (such as small text and other structured fine details) may not be reconstructed accurately.
|
284 |
Verified to have met prescribed NVIDIA quality standards: | Yes
|
285 |
Performance Metrics: | Peak Signal-to-Noise Ratio (PSNR), Structural Similarity (SSIM), Reconstruction Fréchet Video Distance (rFVD), Reconstruction Fréchet Inception Distance (rFID), Latency
|
286 |
Potential Known Risks: | Tokenizer's output can parse all forms of input, including what may be considered toxic, offensive, or indecent.
|