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

candle-examples/examples/segment-anything/README.md

@@ -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 0.6,0.6 --point 0.6,0.55
-    --point-y 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
+The values used for `--point` should be a comma delimited pair of float values.
-are proportional to the image dimension, i.e. use 0.5 for the image center.
+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`.

candle-examples/examples/segment-anything/main.rs

@@ -27,13 +27,9 @@ struct Args {
     #[arg(long)]
     generate_masks: bool,
 
-    /// Comma separated list of x coordinates, between 0 and 1 (0.5 is at the middle of the image).
+    /// List of x,y coordinates, between 0 and 1 (0.5 is at the middle of the image).
-    #[arg(long, use_value_delimiter = true)]
+    #[arg(long)]
-    point_x: Vec<f64>,
+    point: Vec<String>,
-
-    /// Comma separated list of y coordinate, between 0 and 1 (0.5 is at the middle of the image).
-    #[arg(long, use_value_delimiter = true)]
-    point_y: Vec<f64>,
 
     /// The detection threshold for the mask, 0 is the default value, negative values mean a larger
     /// mask, positive makes the mask more selective.
@@ -111,14 +107,18 @@ pub fn main() -> anyhow::Result<()> {
             )?;
         }
     } else {
-        if args.point_x.len() != args.point_y.len() {
+        let points = args
-            anyhow::bail!(
+            .point
-                "number of x coordinates unequal to the number of y coordinates: {} v.s. {}",
+            .iter()
-                args.point_x.len(),
+            .map(|point| {
-                args.point_y.len()
+                use std::str::FromStr;
-            );
+                let xy = point.split(',').collect::<Vec<_>>();
+                if xy.len() != 2 {
+                    anyhow::bail!("expected format for points is 0.4,0.2")
                 }
-        let points: Vec<(f64, f64)> = args.point_x.into_iter().zip(args.point_y).collect();
+                Ok((f64::from_str(xy[0])?, f64::from_str(xy[1])?))
+            })
+            .collect::<anyhow::Result<Vec<_>>>()?;
         let start_time = std::time::Instant::now();
         let (mask, iou_predictions) = sam.forward(&image, &points, false)?;
         println!(

candle-transformers/src/models/segment_anything/sam.rs

@@ -163,14 +163,15 @@ impl Sam {
             None
         } else {
             let n_points = points.len();
-            let mut coords = vec![];
+            let xys = points
-            points.iter().for_each(|(x, y)| {
+                .iter()
+                .flat_map(|(x, y)| {
                     let x = (*x as f32) * (original_w as f32);
                     let y = (*y as f32) * (original_h as f32);
-                coords.push(x);
+                    [x, y]
-                coords.push(y);
+                })
-            });
+                .collect::<Vec<_>>();
-            let points = Tensor::from_vec(coords, (n_points, 1, 2), img_embeddings.device())?;
+            let points = Tensor::from_vec(xys, (n_points, 1, 2), img_embeddings.device())?;
             let labels = Tensor::ones((n_points, 1), DType::F32, img_embeddings.device())?;
             Some((points, labels))
         };