Comments (2)
Hi, for a multi-object dataset, you could simply modify the dataloader to sample one mask each time randomly.
Or you could load all masks and use them as prompts. Then we can change the prompt input to multi-mask, multi-box, multi-points. And the multi-mask output can be used to calculate loss with input multi GT mask.
The latter requires a larger code change and would work better.
from sam-hq.
Thanks for your reply!
Another question is how do I design the loss in this case?
from sam-hq.
Related Issues (20)
- The ap of lvis measured by the public model is inconsistent with the ap in the paper. HOT 1
- How can I test the model HOT 1
- The purpose of self.embedding_maskfeature HOT 3
- About the number of detect boxes HOT 2
- Can I have a more specific using example (code snippet) in README.md? HOT 5
- Question about Ablation study HOT 4
- RuntimeError: Unexpected error from cudaGetDeviceCount(). Did you run some cuda functions before calling NumCudaDevices() that might have already set an error? Error 804: forward compatibility was attempted on non supported HW
- Can we add support for transformers? HOT 3
- RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 1 but got size 13 for tensor number 1 in the list. HOT 1
- Training problem: During the training of sam-hq, the iou output of the val set is very high, 0.98; but in eval mode, the iou of the val set is only 0.48
- Segment Anything CPP Wrapper for macOS
- Can sam-hq perform breakpoint training: continue training on the already trained sam-hq, or continue training on the official sam-hq weights HOT 2
- Alternative implementation in Refiners
- About interm_embeddings
- Runtime error in Colab HOT 1
- How to train vit_tiny (Light HQ-SAM for real-time need): ViT-Tiny HQ-SAM model?
- Why can't we achieve the demonstration effect? There are three tennis rackets in the incoming image, and only one mask is returned
- Visualization of Figure 6
- can't we get candidates as original sam?
- About the issue of positive and negative sample pairs HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from sam-hq.