Use a single flag for the point argument. (#958)

This commit is contained in:
Laurent Mazare
2023-09-25 12:53:24 +01:00
committed by GitHub
parent 7f2bbcf746
commit a36d883254
3 changed files with 31 additions and 31 deletions

View File

@ -16,30 +16,29 @@ based on [MobileSAM](https://github.com/ChaoningZhang/MobileSAM).
cargo run --example segment-anything --release -- \
--image candle-examples/examples/yolo-v8/assets/bike.jpg
--use-tiny
--point-x 0.6,0.6
--point-y 0.6,0.55
--point 0.6,0.6 --point 0.6,0.55
```
Running this command generates a `sam_merged.jpg` file containing the original
image with a blue overlay of the selected mask. The red dots represent the prompt
specified by `--point-x 0.6,0.6 --point-y 0.6,0.55`, this prompt is assumed to be part
specified by `--point 0.6,0.6 --point 0.6,0.55`, this prompt is assumed to be part
of the target mask.
The values used for `--point-x` and `--point-y` should be between 0 and 1 and
are proportional to the image dimension, i.e. use 0.5 for the image center.
The values used for `--point` should be a comma delimited pair of float values.
They are proportional to the image dimension, i.e. use 0.5 for the image center.
Original image:
![Leading group, Giro d'Italia 2021](../yolo-v8/assets/bike.jpg)
Segment results by prompting with a single point `--point-x 0.6 --point-y 0.55`:
Segment results by prompting with a single point `--point 0.6,0.55`:
![Leading group, Giro d'Italia 2021](./assets/single_pt_prompt.jpg)
Segment results by prompting with multiple points `--point-x 0.6,0.6 --point-y 0.6,0.55`:
Segment results by prompting with multiple points `--point 0.6,0.6 --point 0.6,0.55`:
![Leading group, Giro d'Italia 2021](./assets/two_pt_prompt.jpg)
### Command-line flags
- `--use-tiny`: use the TinyViT based MobileSAM backbone rather than the default
one.
- `--point-x`, `--point-y`: specifies the location of the target points.
- `--point`: specifies the location of the target points.
- `--threshold`: sets the threshold value to be part of the mask, a negative
value results in a larger mask and can be specified via `--threshold=-1.2`.