feat(rfdetr): add object detection API (#5923)

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

core/http/endpoints/localai/detection.go

@@ -0,0 +1,59 @@
+package localai
+
+import (
+	"github.com/gofiber/fiber/v2"
+	"github.com/mudler/LocalAI/core/backend"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/http/middleware"
+	"github.com/mudler/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/pkg/utils"
+	"github.com/rs/zerolog/log"
+)
+
+// DetectionEndpoint is the LocalAI Detection endpoint https://localai.io/docs/api-reference/detection
+// @Summary Detects objects in the input image.
+// @Param request body schema.DetectionRequest true "query params"
+// @Success 200 {object} schema.DetectionResponse "Response"
+// @Router /v1/detection [post]
+func DetectionEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
+	return func(c *fiber.Ctx) error {
+
+		input, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_LOCALAI_REQUEST).(*schema.DetectionRequest)
+		if !ok || input.Model == "" {
+			return fiber.ErrBadRequest
+		}
+
+		cfg, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_MODEL_CONFIG).(*config.BackendConfig)
+		if !ok || cfg == nil {
+			return fiber.ErrBadRequest
+		}
+
+		log.Debug().Str("image", input.Image).Str("modelFile", "modelFile").Str("backend", cfg.Backend).Msg("Detection")
+
+		image, err := utils.GetContentURIAsBase64(input.Image)
+		if err != nil {
+			return err
+		}
+
+		res, err := backend.Detection(image, ml, appConfig, *cfg)
+		if err != nil {
+			return err
+		}
+
+		response := schema.DetectionResponse{
+			Detections: make([]schema.Detection, len(res.Detections)),
+		}
+		for i, detection := range res.Detections {
+			response.Detections[i] = schema.Detection{
+				X:         detection.X,
+				Y:         detection.Y,
+				Width:     detection.Width,
+				Height:    detection.Height,
+				ClassName: detection.ClassName,
+			}
+		}
+
+		return c.JSON(response)
+	}
+}

core/http/routes/localai.go

@@ -41,6 +41,11 @@ func RegisterLocalAIRoutes(router *fiber.App,
 		router.Get("/backends/jobs/:uuid", backendGalleryEndpointService.GetOpStatusEndpoint())
 	}
 
+	router.Post("/v1/detection",
+		requestExtractor.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_DETECTION)),
+		requestExtractor.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.DetectionRequest) }),
+		localai.DetectionEndpoint(cl, ml, appConfig))
+
 	router.Post("/tts",
 		requestExtractor.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_TTS)),
 		requestExtractor.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.TTSRequest) }),

core/http/views/backends.html

@@ -90,6 +90,14 @@
                         hx-indicator=".htmx-indicator">
                         <i class="fas fa-headphones mr-2"></i>Whisper
                     </button>
+                    <button hx-post="browse/search/backends" 
+                        class="inline-flex items-center rounded-full px-4 py-2 text-sm font-medium bg-red-900/60 text-red-200 border border-red-700/50 hover:bg-red-800 transition duration-200 ease-in-out"
+                        hx-target="#search-results" 
+                        hx-vals='{"search": "object-detection"}'
+                        onclick="hidePagination()"
+                        hx-indicator=".htmx-indicator">
+                        <i class="fas fa-eye mr-2"></i>Object detection
+                    </button>
                 </div>
             </div>
         </div>

core/http/views/models.html

@@ -115,6 +115,14 @@
                         hx-indicator=".htmx-indicator">
                         <i class="fas fa-headphones mr-2"></i>Audio transcription
                     </button>
+                    <button hx-post="browse/search/models"
+                        class="inline-flex items-center rounded-full px-4 py-2 text-sm font-medium bg-red-900/60 text-red-200 border border-red-700/50 hover:bg-red-800 transition duration-200 ease-in-out"
+                        hx-target="#search-results" 
+                        hx-vals='{"search": "object-detection"}'
+                        onclick="hidePagination()"
+                        hx-indicator=".htmx-indicator">
+                        <i class="fas fa-eye mr-2"></i>Object detection
+                    </button>
                 </div>
             </div>