IDEA-Research/Grounded-SAM-2
Grounded SAM 2: Ground and Track Anything in Videos with Grounding DINO, Florence-2 and SAM 2
Jupyter NotebookApache-2.0
Issues
- 7
Docker GPU Issues
#56 opened by bgiffo96 - 2
- 4
Unexpected label during prediction
#66 opened by yuerout - 4
run florence2 error
#55 opened by yuntao229 - 1
Train GroundedSAM2 with custom dataset
#63 opened by Aseni95 - 1
- 2
When new category of objects occur in the later frame, which are not prompted in the first frame, are there some good methods to tracking the new objects?
#58 opened by RongkunYang - 3
RuntimeError("No points are provided; please add points first") during reverse tracking
#53 opened by SJP2022 - 5
- 5
Assertion Error
#26 opened by Masrur02 - 0
Object class_name BUG?
#54 opened by Li-jinnan - 3
The GPU memory is insufficient?
#51 opened by Li-jinnan - 1
Output class names return ungiven classes
#50 opened by CYBruce - 2
Missing Python libraries in the Dockerfile
#46 opened by yassineAlouini - 1
- 1
Correct masks but wrong classes over time
#49 opened by rmarcuzzi - 7
- 3
If our video does not initially contain the object we want to detect, how should we modify the code?
#45 opened by Zhujx-zjjx - 2
Can CUDA 12.2 be used for the Grounded-SAM-2
#33 opened by Masrur02 - 1
- 2
Support image prompt
#29 opened by wentao-uw - 3
SAM 2 Florence-2 pipeline image
#31 opened by spped2000 - 3
Making SAM 2 run 2x faster
#37 opened by mvoodarla - 4
How to track new objects?
#19 opened by melodyhappy - 2
Fine tuning?
#41 opened by GXcells - 2
- 5
Real-Time Inference
#16 opened by Henistein - 1
Florence-2 vs Grounding DINO + SAM2
#34 opened by radames - 3
Why there is a assert?
#39 opened by numb89757 - 2
- 6
ImportError: flash_attn not found when running grounded_sam2_florence2_image_demo.py
#35 opened by XiaowenZhang-kuku - 1
grounded_sam2_tracking_demo_with_continuous_id
#32 opened by wjpoom - 1
Is there a parameter in the code that can control the tracking accuracy of the video predictor and reduce the false positives (FP)?
#22 opened by melodyhappy - 3
prompt type for video issue
#28 opened by ZhangT-tech - 7
- 1
ComfyUI surpport?
#11 opened by Pythonpa - 3
Predict from a video file
#25 opened by Masrur02 - 3
GroundingDINO Import Error
#24 opened by cremebrule - 2
Does Florence2 can predict multiple object in Open Vocabulary Detection and Segmentation mode?
#23 opened by LiuShihHung - 3
How about 3D version?
#20 opened by StarsTesla - 4
OOM occurs when adding certain box prompts.
#21 opened by MrIsland - 8
CUDA out of memory
#13 opened by Masrur02 - 3
How to divide the road lane
#15 opened by SUN-zx - 1
Tracking ID varied
#14 opened by TheMattBin - 3
- 3
ValueError: mask must be a 3D np.ndarray with shape (1, H, W), but got shape (1, 1, 1024, 768) when using prompt "shirt" in Grounded SAM 2
#5 opened by sahal-786 - 5
请问这个报错怎么解决呢 n, h, w = masks.shape ValueError: too many values to unpack (expected 3)
#4 opened by shnulailai - 4
About SAM2 Prompt
#1 opened by ShuoShenDe - 3
Perhaps directly using mask prompt or box prompt in SAM 2 (instead of Uniform Point Sampling)?
#3 opened by ronghanghu