📝 Update Mask Generation Task (#432)

Added about.md for Zero-shot-object-detection, text-to-3d and
mask-generation. Also added use cases for them all.

---------

Co-authored-by: Pedro Cuenca <[email protected]>
Co-authored-by: Eliott C. <[email protected]>
Co-authored-by: coyotte508 <[email protected]>

Author: 3 people

packages/tasks/.prettierignore

@@ -1,4 +1,5 @@
 pnpm-lock.yaml
 # In order to avoid code samples to have tabs, they don't display well on npm
 README.md
-dist
+dist
+.tshy

packages/tasks/src/tasks/mask-generation/about.md

@@ -12,6 +12,16 @@ Generating masks can facilitate learning, especially in semi or unsupervised lea
 
 For applications where humans are in the loop, masks highlight certain regions of images for humans to validate.
 
+### Medical Imaging
+
+Mask generation models are used in medical imaging to aid in segmenting and analyzing specific regions.
+
+### Autonomous Vehicles
+
+Mask generation models are used to create segments and masks for obstacles and other objects in view.
+
+This page was made possible thanks to the efforts of [Raj Aryan](https://huggingface.co/thatrajaryan) and other contributors.
+
 ## Task Variants
 
 ### Segmentation

packages/tasks/src/tasks/mask-generation/data.ts

@@ -1,7 +1,16 @@
 import type { TaskDataCustom } from "../index.js";
 
 const taskData: TaskDataCustom = {
-	datasets: [],
+	datasets: [
+		{
+			description: "Widely used benchmark dataset for multiple Vision tasks.",
+			id: "merve/coco2017",
+		},
+		{
+			description: "Medical Imaging dataset of the Human Brain for segmentation and mask generating tasks",
+			id: "rocky93/BraTS_segmentation",
+		},
+	],
 	demo: {
 		inputs: [
 			{
@@ -16,7 +25,12 @@ const taskData: TaskDataCustom = {
 			},
 		],
 	},
-	metrics: [],
+	metrics: [
+		{
+			description: "IoU is used to measure the overlap between predicted mask and the ground truth mask.",
+			id: "Intersection over Union (IoU)",
+		},
+	],
 	models: [
 		{
 			description: "Small yet powerful mask generation model.",