[Enhancement] Add Visual Grounding / Detection by Image Input

[Interpause]

DINO-X-MCP/src/dino-x/client.ts

Lines 186 to 201 in 83be2e9

	async detectObjectsByText(
	imageFileUri: string,
	textPrompt: string,
	includeDescription: boolean
	): Promise<DetectionResult> {
	return this.performDetection(imageFileUri, includeDescription, {
	model: "DINO-X-1.0",
	prompt: {
	type: "text",
	text: textPrompt
	},
	targets: ["bbox"],
	bbox_threshold: 0.25,
	iou_threshold: 0.8
	});
	}

Hi, DINO-X supports visual prompts for visual grounding-based object detection right? Would it be possible to add a tool that lets the agent give a local file URL/HTTPS URL to use as the detection prompt? I think just need to modify the function above right? Thanks.

[cefeng06]

Hi! Thanks for your question.

Currently, the DINO-X API doesn’t support visual prompts for grounding or detection yet, so passing a local or HTTPS image as a prompt isn’t available.