Fixes (#18139)

* Catch error and show toast when failing to delete review items

* i18n keys

* add link to speed estimation docs in zone edit pane

* Implement reset of tracked object update for each camera

* Cleanup

* register mqtt callbacks for toggling alerts and detections

* clarify snapshots docs

* clarify semantic search reindexing

* add ukrainian

* adjust date granularity for last recording time

The api endpoint only returns granularity down to the day

* Add amd hardware

* fix crash in face library on initial start after enabling

* Fix recordings view for mobile landscape

The events view incorrectly was displaying two columns on landscape view and it only took up 20% of the screen width. Additionally, in landscape view the timeline was too wide (especially on iPads of various screen sizes) and would overlap the main video

* face rec overfitting instructions

* Clarify

* face docs

* clarify

* clarify

---------

Co-authored-by: Nicolas Mowen <nickmowen213@gmail.com>

docs/docs/configuration/face_recognition.md

@@ -137,6 +137,15 @@ This can happen for a few different reasons, but this is usually an indicator th
 - When you provide images with different poses, lighting, and expressions, the algorithm extracts features that are consistent across those variations.
 - By training on a diverse set of images, the algorithm becomes less sensitive to minor variations and noise in the input image.
 
+Review your face collections and remove most of the unclear or low-quality images. Then, use the **Reprocess** button on each face in the **Train** tab to evaluate how the changes affect recognition scores.
+
+Avoid training on images that already score highly, as this can lead to over-fitting. Instead, focus on relatively clear images that score lower - ideally with different lighting, angles, and conditions—to help the model generalize more effectively.
+
+### Frigate misidentified a face. Can I tell it that a face is "not" a specific person?
+
+No, face recognition does not support negative training (i.e., explicitly telling it who someone is _not_). Instead, the best approach is to improve the training data by using a more diverse and representative set of images for each person.
+For more guidance, refer to the section above on improving recognition accuracy.
+
 ### I see scores above the threshold in the train tab, but a sub label wasn't assigned?
 
 The Frigate considers the recognition scores across all recognition attempts for each person object. The scores are continually weighted based on the area of the face, and a sub label will only be assigned to person if a person is confidently recognized consistently. This avoids cases where a single high confidence recognition would throw off the results.

docs/docs/configuration/semantic_search.md

@@ -19,7 +19,7 @@ For best performance, 16GB or more of RAM and a dedicated GPU are recommended.
 
 ## Configuration
 
-Semantic Search is disabled by default, and must be enabled in your config file or in the UI's Settings page before it can be used. Semantic Search is a global configuration setting.
+Semantic Search is disabled by default, and must be enabled in your config file or in the UI's Classification Settings page before it can be used. Semantic Search is a global configuration setting.
 
 ```yaml
 semantic_search:
@@ -29,9 +29,9 @@ semantic_search:
 
 :::tip
 
-The embeddings database can be re-indexed from the existing tracked objects in your database by adding `reindex: True` to your `semantic_search` configuration or by toggling the switch on the Search Settings page in the UI and restarting Frigate. Depending on the number of tracked objects you have, it can take a long while to complete and may max out your CPU while indexing. Make sure to turn the UI's switch off or set the config back to `False` before restarting Frigate again.
+The embeddings database can be re-indexed from the existing tracked objects in your database by pressing the "Reindex" button in the Classification Settings in the UI or by adding `reindex: True` to your `semantic_search` configuration and restarting Frigate. Depending on the number of tracked objects you have, it can take a long while to complete and may max out your CPU while indexing.
 
-If you are enabling Semantic Search for the first time, be advised that Frigate does not automatically index older tracked objects. You will need to enable the `reindex` feature in order to do that.
+If you are enabling Semantic Search for the first time, be advised that Frigate does not automatically index older tracked objects. You will need to reindex as described above.
 
 :::
 
@@ -72,7 +72,7 @@ For most users, especially native English speakers, the V1 model remains the rec
 
 :::note
 
-Switching between V1 and V2 requires reindexing your embeddings. To do this, set `reindex: True` in your Semantic Search configuration and restart Frigate. The embeddings from V1 and V2 are incompatible, and failing to reindex will result in incorrect search results.
+Switching between V1 and V2 requires reindexing your embeddings. The embeddings from V1 and V2 are incompatible, and failing to reindex will result in incorrect search results.
 
 :::
 

docs/docs/configuration/snapshots.md

@@ -5,7 +5,7 @@ title: Snapshots
 
 Frigate can save a snapshot image to `/media/frigate/clips` for each object that is detected named as `<camera>-<id>.jpg`. They are also accessible [via the api](../integrations/api/event-snapshot-events-event-id-snapshot-jpg-get.api.mdx)
 
-For users with Frigate+ enabled, snapshots are accessible in the UI in the Frigate+ pane to allow for quick submission to the Frigate+ service.
+Snapshots are accessible in the UI in the Explore pane. This allows for quick submission to the Frigate+ service.
 
 To only save snapshots for objects that enter a specific zone, [see the zone docs](./zones.md#restricting-snapshots-to-specific-zones)
 

docs/docs/frigate/hardware.md

@@ -143,9 +143,10 @@ Inference speeds will vary greatly depending on the GPU and the model used.
 
 With the [rocm](../configuration/object_detectors.md#amdrocm-gpu-detector) detector Frigate can take advantage of many discrete AMD GPUs.
 
-| Name     | YOLOv9 Inference Time | YOLO-NAS Inference Time   |
-| -------- | --------------------- | ------------------------- |
-| AMD 780M | ~ 14 ms               | 320: ~ 30 ms 640: ~ 60 ms |
+| Name      | YOLOv9 Inference Time | YOLO-NAS Inference Time   |
+| --------- | --------------------- | ------------------------- |
+| AMD 780M  | ~ 14 ms               | 320: ~ 30 ms 640: ~ 60 ms |
+| AMD 8700G |                       | 320: ~ 20 ms 640: ~ 40 ms |
 
 ## Community Supported Detectors
 

frigate/comms/mqtt.py

@@ -213,6 +213,8 @@ class MqttClient(Communicator):  # type: ignore[misc]
             "motion_contour_area",
             "birdseye",
             "birdseye_mode",
+            "review_alerts",
+            "review_detections",
         ]
 
         for name in self.config.cameras.keys():

frigate/data_processing/common/license_plate/mixin.py

@@ -1552,6 +1552,12 @@ class LicensePlateProcessingMixin:
                 (base64.b64encode(encoded_img).decode("ASCII"), id, camera),
             )
 
+        if id not in self.detected_license_plates:
+            if camera not in self.camera_current_cars:
+                self.camera_current_cars[camera] = []
+
+            self.camera_current_cars[camera].append(id)
+
         self.detected_license_plates[id] = {
             "plate": top_plate,
             "char_confidences": top_char_confidences,
@@ -1564,7 +1570,7 @@ class LicensePlateProcessingMixin:
     def handle_request(self, topic, request_data) -> dict[str, any] | None:
         return
 
-    def expire_object(self, object_id: str):
+    def expire_object(self, object_id: str, camera: str):
         if object_id in self.detected_license_plates:
             self.detected_license_plates.pop(object_id)
 

frigate/data_processing/real_time/api.py

@@ -50,10 +50,11 @@ class RealTimeProcessorApi(ABC):
         pass
 
     @abstractmethod
-    def expire_object(self, object_id: str) -> None:
+    def expire_object(self, object_id: str, camera: str) -> None:
         """Handle objects that are no longer detected.
         Args:
             object_id (str): id of object that is no longer detected.
+            camera (str): name of camera that object was detected on.
 
         Returns:
             None.

frigate/data_processing/real_time/bird.py

@@ -152,6 +152,6 @@ class BirdRealTimeProcessor(RealTimeProcessorApi):
     def handle_request(self, topic, request_data):
         return None
 
-    def expire_object(self, object_id):
+    def expire_object(self, object_id, camera):
         if object_id in self.detected_birds:
             self.detected_birds.pop(object_id)

frigate/data_processing/real_time/face.py

@@ -54,6 +54,7 @@ class FaceRealTimeProcessor(RealTimeProcessorApi):
         self.face_detector: cv2.FaceDetectorYN = None
         self.requires_face_detection = "face" not in self.config.objects.all_objects
         self.person_face_history: dict[str, list[tuple[str, float, int]]] = {}
+        self.camera_current_people: dict[str, list[str]] = {}
         self.recognizer: FaceRecognizer | None = None
         self.faces_per_second = EventsPerSecond()
         self.inference_speed = InferenceSpeed(self.metrics.face_rec_speed)
@@ -282,9 +283,13 @@ class FaceRealTimeProcessor(RealTimeProcessorApi):
         if id not in self.person_face_history:
             self.person_face_history[id] = []
 
+            if camera not in self.camera_current_people:
+                self.camera_current_people[camera] = []
+
         self.person_face_history[id].append(
             (sub_label, score, face_frame.shape[0] * face_frame.shape[1])
         )
+        self.camera_current_people[camera].append(id)
         (weighted_sub_label, weighted_score) = self.weighted_average(
             self.person_face_history[id]
         )
@@ -420,10 +425,25 @@ class FaceRealTimeProcessor(RealTimeProcessorApi):
                 )
                 shutil.move(current_file, new_file)
 
-    def expire_object(self, object_id: str):
+    def expire_object(self, object_id: str, camera: str):
         if object_id in self.person_face_history:
             self.person_face_history.pop(object_id)
 
+            if object_id in self.camera_current_people.get(camera, []):
+                self.camera_current_people[camera].remove(object_id)
+
+                if len(self.camera_current_people[camera]) == 0:
+                    self.requestor.send_data(
+                        "tracked_object_update",
+                        json.dumps(
+                            {
+                                "type": TrackedObjectUpdateTypesEnum.face,
+                                "name": None,
+                                "camera": camera,
+                            }
+                        ),
+                    )
+
     def weighted_average(
         self, results_list: list[tuple[str, float, int]], max_weight: int = 4000
     ):

frigate/data_processing/real_time/license_plate.py

@@ -1,5 +1,6 @@
 """Handle processing images for face detection and recognition."""
 
+import json
 import logging
 
 import numpy as np
@@ -13,6 +14,7 @@ from frigate.data_processing.common.license_plate.mixin import (
 from frigate.data_processing.common.license_plate.model import (
     LicensePlateModelRunner,
 )
+from frigate.types import TrackedObjectUpdateTypesEnum
 
 from ..types import DataProcessorMetrics
 from .api import RealTimeProcessorApi
@@ -36,6 +38,7 @@ class LicensePlateRealTimeProcessor(LicensePlateProcessingMixin, RealTimeProcess
         self.lpr_config = config.lpr
         self.config = config
         self.sub_label_publisher = sub_label_publisher
+        self.camera_current_cars: dict[str, list[str]] = {}
         super().__init__(config, metrics)
 
     def process_frame(
@@ -50,6 +53,22 @@ class LicensePlateRealTimeProcessor(LicensePlateProcessingMixin, RealTimeProcess
     def handle_request(self, topic, request_data) -> dict[str, any] | None:
         return
 
-    def expire_object(self, object_id: str):
+    def expire_object(self, object_id: str, camera: str):
         if object_id in self.detected_license_plates:
             self.detected_license_plates.pop(object_id)
+
+            if object_id in self.camera_current_cars.get(camera, []):
+                self.camera_current_cars[camera].remove(object_id)
+
+                if len(self.camera_current_cars[camera]) == 0:
+                    self.requestor.send_data(
+                        "tracked_object_update",
+                        json.dumps(
+                            {
+                                "type": TrackedObjectUpdateTypesEnum.lpr,
+                                "name": None,
+                                "plate": None,
+                                "camera": camera,
+                            }
+                        ),
+                    )

frigate/embeddings/maintainer.py

@@ -359,7 +359,7 @@ class EmbeddingMaintainer(threading.Thread):
 
             # expire in realtime processors
             for processor in self.realtime_processors:
-                processor.expire_object(event_id)
+                processor.expire_object(event_id, camera)
 
             if updated_db:
                 try:

web/public/locales/en/common.json

@@ -60,6 +60,10 @@
       "12hour": "MMM d, h:mm aaa",
       "24hour": "MMM d, HH:mm"
     },
+    "formattedTimestampMonthDayYear": {
+      "12hour": "MMM d, yyyy",
+      "24hour": "MMM d, yyyy"
+    },
     "formattedTimestampMonthDayYearHourMinute": {
       "12hour": "MMM d yyyy, h:mm aaa",
       "24hour": "MMM d yyyy, HH:mm"

web/public/locales/en/components/dialog.json

@@ -98,6 +98,10 @@
       "title": "Confirm Delete",
       "desc": {
         "selected": "Are you sure you want to delete all recorded video associated with this review item?<br /><br />Hold the <em>Shift</em> key to bypass this dialog in the future."
+      },
+      "toast": {
+        "success": "Video footage associated with the selected review items has been deleted successfully.",
+        "error": "Failed to delete: {{error}}"
       }
     },
     "button": {

web/public/locales/en/views/faceLibrary.json

@@ -68,6 +68,7 @@
     "dropInstructions": "Drag and drop an image here, or click to select",
     "maxSize": "Max size: {{size}}MB"
   },
+  "nofaces": "No faces available",
   "readTheDocs": "Read the documentation",
   "trainFaceAs": "Train Face as:",
   "trainFace": "Train Face",

web/public/locales/en/views/settings.json

@@ -268,7 +268,8 @@
       "allObjects": "All Objects",
       "speedEstimation": {
         "title": "Speed Estimation",
-        "desc": "Enable speed estimation for objects in this zone. The zone must have exactly 4 points."
+        "desc": "Enable speed estimation for objects in this zone. The zone must have exactly 4 points.",
+        "docs": "Read the documentation"
       },
       "speedThreshold": {
         "title": "Speed Threshold ({{unit}})",

web/src/components/filter/ReviewActionGroup.tsx

@@ -17,6 +17,7 @@ import {
 } from "../ui/alert-dialog";
 import useKeyboardListener from "@/hooks/use-keyboard-listener";
 import { Trans, useTranslation } from "react-i18next";
+import { toast } from "sonner";
 
 type ReviewActionGroupProps = {
   selectedReviews: string[];
@@ -41,11 +42,33 @@ export default function ReviewActionGroup({
     pullLatestData();
   }, [selectedReviews, setSelectedReviews, pullLatestData]);
 
-  const onDelete = useCallback(async () => {
-    await axios.post(`reviews/delete`, { ids: selectedReviews });
-    setSelectedReviews([]);
-    pullLatestData();
-  }, [selectedReviews, setSelectedReviews, pullLatestData]);
+  const onDelete = useCallback(() => {
+    axios
+      .post(`reviews/delete`, { ids: selectedReviews })
+      .then((resp) => {
+        if (resp.status === 200) {
+          toast.success(t("recording.confirmDelete.toast.success"), {
+            position: "top-center",
+          });
+          setSelectedReviews([]);
+          pullLatestData();
+        }
+      })
+      .catch((error) => {
+        const errorMessage =
+          error.response?.data?.message ||
+          error.response?.data?.detail ||
+          "Unknown error";
+        toast.error(
+          t("recording.confirmDelete.toast.error", {
+            error: errorMessage,
+          }),
+          {
+            position: "top-center",
+          },
+        );
+      });
+  }, [selectedReviews, setSelectedReviews, pullLatestData, t]);
 
   const [deleteDialogOpen, setDeleteDialogOpen] = useState(false);
   const [bypassDialog, setBypassDialog] = useState(false);

web/src/components/settings/ZoneEditPane.tsx

@@ -30,6 +30,8 @@ import { flattenPoints, interpolatePoints } from "@/utils/canvasUtil";
 import ActivityIndicator from "../indicators/activity-indicator";
 import { getAttributeLabels } from "@/utils/iconUtil";
 import { Trans, useTranslation } from "react-i18next";
+import { Link } from "react-router-dom";
+import { LuExternalLink } from "react-icons/lu";
 
 type ZoneEditPaneProps = {
   polygons?: Polygon[];
@@ -669,6 +671,17 @@ export default function ZoneEditPane({
                 </div>
                 <FormDescription>
                   {t("masksAndZones.zones.speedEstimation.desc")}
+                  <div className="mt-2 flex items-center text-primary">
+                    <Link
+                      to="https://docs.frigate.video/configuration/zones#speed-estimation"
+                      target="_blank"
+                      rel="noopener noreferrer"
+                      className="inline"
+                    >
+                      {t("masksAndZones.zones.speedEstimation.docs")}
+                      <LuExternalLink className="ml-2 inline-flex size-3" />
+                    </Link>
+                  </div>
                 </FormDescription>
                 <FormMessage />
               </FormItem>

web/src/lib/const.ts

@@ -11,4 +11,5 @@ export const supportedLanguageKeys = [
   "zh-CN",
   "yue-Hant",
   "ru",
+  "uk",
 ];

web/src/pages/FaceLibrary.tsx

@@ -390,10 +390,10 @@ export default function FaceLibrary() {
           </div>
         )}
       </div>
-      {pageToggle && faceImages.length === 0 && pageToggle !== "train" ? (
+      {pageToggle && faceImages?.length === 0 && pageToggle !== "train" ? (
         <div className="absolute left-1/2 top-1/2 flex -translate-x-1/2 -translate-y-1/2 flex-col items-center justify-center text-center">
           <LuFolderCheck className="size-16" />
-          No faces available
+          {t("nofaces")}
         </div>
       ) : (
         pageToggle &&

web/src/views/recording/RecordingView.tsx

@@ -819,7 +819,7 @@ function Timeline({
       className={`${
         isDesktop
           ? `${timelineType == "timeline" ? "w-[100px]" : "w-60"} no-scrollbar overflow-y-auto`
-          : "overflow-hidden portrait:flex-grow landscape:w-[20%]"
+          : `overflow-hidden portrait:flex-grow ${timelineType == "timeline" ? "landscape:w-[100px]" : "landscape:w-[175px]"} `
       } relative`}
     >
       <div className="pointer-events-none absolute inset-x-0 top-0 z-20 h-[30px] w-full bg-gradient-to-b from-secondary to-transparent"></div>
@@ -855,7 +855,7 @@ function Timeline({
           <div
             className={cn(
               "scrollbar-container grid h-auto grid-cols-1 gap-4 overflow-auto p-4",
-              isMobile && "sm:grid-cols-2",
+              isMobile && "sm:portrait:grid-cols-2",
             )}
           >
             {mainCameraReviewItems.length === 0 ? (

web/src/views/system/StorageMetrics.tsx

@@ -71,7 +71,7 @@ export default function StorageMetrics({
 
   const timeFormat = config?.ui.time_format === "24hour" ? "24hour" : "12hour";
   const format = useMemo(() => {
-    return t(`time.formattedTimestampMonthDayYearHourMinute.${timeFormat}`, {
+    return t(`time.formattedTimestampMonthDayYear.${timeFormat}`, {
       ns: "common",
     });
   }, [t, timeFormat]);