nicolabs/Open-Sora
1
0
Fork 0
mirror of https://github.com/hpcaitech/Open-Sora.git synced 2026-05-21 11:59:01 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

update pllava readme

Browse source
This commit is contained in:
Tom Young 2024-06-17 06:02:43 +00:00
parent b2670f57fa
commit f3b8c861e5
1 changed files with 2 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

5
tools/caption/README.md
View file

@ -62,19 +62,18 @@ nohup python caption_pllava.py \
### PLLaVA VS LLaVA
In our previous releases, we used [LLaVA](#llava-captioning) for video captioning.
Qualitatively speaking, we observe that PLLaVA has a somewhat higher chance of accurately capture the details in the video.
Qualitatively speaking, we observe that PLLaVA has a somewhat higher chance of accurately capture the details in the video than LLaVA. See below for their comparison on a video sample.
<!-- <img src="../../assets/readme/llava_vs_pllava_sample.gif" width="300" height="200" alt="LLaVA vs PLLaVA"> -->
<figure>
<img src="../../assets/readme/llava_vs_pllava_sample.gif" width="300" height="200" alt="LLaVA vs PLLaVA">
<figcaption>Sample video for the comparison below</figcaption>
</figure>
| LLaVA | PLLaVA |
|----------|----------|
| The video is a close-up shot of two gold wedding rings. The rings are placed on a <span style="color: green;">black surface</span>, casting a soft shadow beneath them. The rings are positioned in such a way that <span style="color: red;">they are facing each other, creating a mirror image effect.</span> The rings are identical in size and design, suggesting they are a pair. The lighting in the video is soft and diffused, highlighting the gold color of the rings and creating a warm and inviting atmosphere. The overall style of the video is minimalist and elegant, focusing solely on the rings and their reflection. | The video shows a pair of gold wedding rings on a <span style="color: green;">reflective surface</span>. The rings are placed one on top of the other, with <span style="color: green;">the top ring slightly tilted to the left</span>. The rings have a shiny, metallic finish and are the main focus of the image. The background is a gradient of dark to light gray, providing a neutral backdrop that highlights the rings. There are no texts or other objects in the image. The style of the video is a simple product display with a focus on the rings, likely intended for promotional or sales purposes. The lighting and shadows suggest a soft, even light source, possibly a studio light, which creates a reflective surface beneath the rings.|
| The video is a close-up shot of two gold wedding rings. The rings are placed on a (black surface)✅, casting a soft shadow beneath them. The rings are positioned in such a way that (they are facing each other)❌, creating a mirror image effect. The rings are (identical in size and design)✅, suggesting they are a pair. The lighting in the video is soft and diffused, highlighting the gold color of the rings and creating a warm and inviting atmosphere. The overall style of the video is minimalist and elegant, focusing solely on the rings and their reflection. | The video shows a pair of gold wedding rings on a (reflective surface)✅. The rings are placed one on top of the other, with the top ring slightly tilted to the left. The rings have a (shiny, metallic finish)✅ and are the main focus of the image. The background is a gradient of dark to light gray, providing a neutral backdrop that highlights the rings. There are no texts or other objects in the image. The style of the video is a simple product display with a focus on the rings, likely intended for promotional or sales purposes. The lighting and shadows suggest a soft, even light source, (possibly a studio light)✅, which creates a reflective surface beneath the rings.|
<!-- |Row2Cell1|Row2Cell2| -->