deepseek-ai / deepseek-ai/DeepSeek-VL2
visual grounding problems
Open
- Dominant language
- Python
- Stars
- 5.4k
- Forks
- 1.8k
- PR merge metrics
- No merged PRs in 30d
Description
1. I find it is poor when using VL2, maybe tiny or small version will be better?
2. I wonder its representation format, because somtimes it will return odd number of numbers, like [533, 132, 457]，I don’t know how can it represent a bounding box.
3. Sometimes，it returns nested list(For example: [245,[234,543,542,244]]. What does it mean?
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.