# Pre-built models
In the community space, more than 50+ pre-built models and pipelines are ready for use and deployment. In this section, we describe the background of the models, and its associated labels and use cases.
# Objects
In the Detect block, you can input a comma-separated case-sensitive string with the object names like this. Ordering has no significance in detection process.
object1,object2
# default
This is a model adopted from YOLOv4. The most commonly used objects are person and car. The full body of the person can be detected but not the specific body parts. If you need to detect specific body parts, please consider to use crowd-v2 or pose.
# Labels
The label string is as follows.
airplane,apple,backpack,banana,baseball bat,baseball glove,bear,bed,bench,bicycle,bird,boat,book,bottle,bowl,broccoli,bus,cake,car,carrot,cat,cell phone,chair,clock,couch,cow,cup,dining table,dog,donut,elephant,fire hydrant,fork,frisbee,giraffe,hair drier,handbag,horse,hot dog,keyboard,kite,knife,laptop,microwave,motorcycle,mouse,orange,oven,parking meter,person,pizza,potted plant,refrigerator,remote,sandwich,scissors,sheep,sink,skateboard,skis,snowboard,spoon,sports ball,stop sign,suitcase,surfboard,teddy bear,tennis racket,tie,toaster,toilet,toothbrush,traffic light,train,truck,tv,umbrella,vase,wine glass,zebra
| Object name | Description |
|---|---|
| airplane | Airplane |
| apple | Apple |
| backpack | Backpack |
| banana | Banana |
| baseball bat | Bat |
| baseball glove | Glove |
| bear | Bear |
| bed | Bed |
| bench | Bench |
| bicycle | Bicycle |
| bird | Bird |
| boat | Boat |
| book | Book |
| bottle | Bottle |
| bowl | Bowl |
| broccoli | Broccoli |
| bus | Bus |
| cake | Cake |
| car | Car |
| carrot | Carrot |
| cat | Cat |
| cell phone | Phone |
| chair | Chair |
| clock | Clock |
| couch | Couch |
| cow | Cow |
| cup | Cup |
| dining table | Table |
| dog | Dog |
| donut | Donut |
| elephant | Elephant |
| fire hydrant | Fire hydrant |
| fork | Fork |
| frisbee | Frisbee |
| giraffe | Giraffe |
| hair drier | Drier |
| handbag | Handbag |
| horse | Horse |
| hot dog | Dog |
| keyboard | Keyboard |
| kite | Kite |
| knife | Knife |
| laptop | Laptop |
| microwave | Microwave |
| motorcycle | Motorcycle |
| mouse | Mouse |
| orange | Orange |
| oven | Oven |
| parking meter | Meter |
| person | Person |
| pizza | Pizza |
| potted plant | Plant |
| refrigerator | Refrigerator |
| remote | Remote |
| sandwich | Sandwich |
| scissors | Scissors |
| sheep | Sheep |
| sink | Sink |
| skateboard | Skateboard |
| skis | Skis |
| snowboard | Snowboard |
| spoon | Spoon |
| sports ball | Sports ball |
| stop sign | Sign |
| suitcase | Suitcase |
| surfboard | Surfboard |
| teddy bear | Bear |
| tennis racket | Racket |
| tie | Tie |
| toaster | Toaster |
| toilet | Toilet |
| toothbrush | Toothbrush |
| traffic light | Traffic light |
| train | Train |
| truck | Truck |
| tv | TV |
| umbrella | Umbrella |
| vase | Vase |
| wine glass | Glass |
| zebra | Zebra |
# AnimalClassifier2
This is an Animal Classifier which classifies 100+ different types of animals commonly found in nature including pets, insects, marine life and wildlife species. The model would be useful for performing real-time animal detection. The average precision and recall rate are 92.5% and 88.7% respectively.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
antelope,badger,bald_eagle,bat,bee,beetle,bison,black_bear,boar,bobcat,buffalo,butterfly,canada_lynx,cat,caterpillar,chimpanzee,cockroach,columbian_black_tailed_deer,cougar,cow,coyote,crab,crow,deer,dog,dolphin,donkey,dragonfly,duck,elephant,elk,flamingo,fly,goat,goldfish,goose,gorilla,grasshopper,gray_fox,gray_wolf,grizzly_bear,hamster,hare,hedgehog,hippopotamus,hornbill,horse,hummingbird,hyena,jellyfish,kangaroo,koala,ladybugs,leopard,lion,lizard,lobster,mosquito,moth,mountain_beaver,mouse,nutria,octopus,okapi,orangutan,otter,owl,ox,oyster,panda,parrot,pelecaniformes,penguin,pig,pigeon,porcupine,possum,raccoon,rat,raven,red_fox,reindeer,rhino,rhinoceros,ringtail,sandpiper,sea_lions,seahorse,seal,shark,sheep,snake,sparrow,squid,squirrel,starfish,swan,teddy_bear,tiger,turkey,turtle,virginia_opossum,whale,wolf,wombat,woodpecker,zebra
| Object name | Description |
|---|---|
| antelope | Antelope |
| badger | Badger |
| bald_eagle | Bald eagle |
| bat | Bat |
| bee | Bee |
| beetle | Beetle |
| bison | Bison |
| black_bear | Black bear |
| boar | Boar |
| bobcat | Bobcat |
| buffalo | Buffalo |
| butterfly | Butterfly |
| canada_lynx | Canada lynx |
| cat | Cat |
| caterpillar | Caterpillar |
| chimpanzee | Chimpanzee |
| cockroach | Cockroach |
| columbian_black_tailed_deer | Columbian black-tailed deer |
| cougar | Cougar |
| cow | Cow |
| coyote | Coyote |
| crab | Crab |
| crow | Crow |
| deer | Deer |
| dog | Dog |
| dolphin | Dolphin |
| donkey | Donkey |
| dragonfly | Dragonfly |
| duck | Duck |
| elephant | Elephant |
| elk | Elk |
| flamingo | Flamingo |
| fly | Fly |
| goat | Goat |
| goldfish | Goldfish |
| goose | Goose |
| gorilla | Gorilla |
| grasshopper | Grasshopper |
| gray_fox | Gray fox |
| gray_wolf | Gray wolf |
| grizzly_bear | Grizzly bear |
| hamster | Hamster |
| hare | Hare |
| hedgehog | Hedgehog |
| hippopotamus | Hippopotamus |
| hornbill | Hornbill |
| horse | Horse |
| hummingbird | Hummingbird |
| hyena | Hyena |
| jellyfish | Jellyfish |
| kangaroo | Kangaroo |
| koala | Koala |
| ladybugs | Ladybugs |
| leopard | Leopard |
| lion | Lion |
| lizard | Lizard |
| lobster | Lobster |
| mosquito | Mosquito |
| moth | Moth |
| mountain_beaver | Mountain beaver |
| mouse | Mouse |
| nutria | Nutria |
| octopus | Octopus |
| okapi | Okapi |
| orangutan | Orangutan |
| otter | Otter |
| owl | Owl |
| ox | Ox |
| oyster | Oyster |
| panda | Panda |
| parrot | Parrot |
| pelecaniformes | Pelecaniformes |
| penguin | Penguin |
| pig | Pig |
| pigeon | Pigeon |
| porcupine | Porcupine |
| possum | Possum |
| raccoon | Raccoon |
| rat | Rat |
| raven | Raven |
| red_fox | Red fox |
| reindeer | Reindeer |
| rhino | Rhino |
| rhinoceros | Rhinoceros |
| ringtail | Ringtail |
| sandpiper | Sandpiper |
| sea_lions | Sea lions |
| seahorse | Seahorse |
| seal | Seal |
| shark | Shark |
| sheep | Sheep |
| snake | Snake |
| sparrow | Sparrow |
| squid | Squid |
| squirrel | Squirrel |
| starfish | Starfish |
| swan | Swan |
| teddy_bear | Teddy bear |
| tiger | Tiger |
| turkey | Turkey |
| turtle | Turtle |
| virginia_opossum | Virginia opossum |
| whale | Whale |
| wolf | Wolf |
| wombat | Wombat |
| woodpecker | Woodpecker |
| zebra | Zebra |
# AnimalClassifier3
This is an improved version of Animal Classifier which classifies 10 different monkey species as well as 106 other types of pets, insects, marine life and wildlife species. More than 20,000 samples in various location were used for training in the dataset with an average precision and recall rate of 93.4% and 90.4% respectively.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
antelope,antelope,badger,bald_eagle,bald_uakari,bat,bee,beetle,bison,black_bear,black_headed_night_monkey,boar,bobcat,buffalo,butterfly,canada_lynx,cat,caterpillar,chimpanzee,cockroach,columbian_black_tailed_deer,common_squirrel_monkey,cougar,cow,coyote,crab,crow,deer,dog,dolphin,donkey,dragonfly,duck,elephant,elk,flamingo,fly,goat,goldfish,goose,gorilla,grasshopper,gray_fox,gray_wolf,grizzly_ bear,hamster,hare,hedgehog,hippopotamus,hornbill,horse,hummingbird,hyena,japanese_macaque,jellyfish,kangaroo,koala,ladybugs,leopard,lion,lizard,lobster,mantled_howler,mosquito,moth,mountain_beaver,mouse,nilgiri_langur,nutria,octopus,okapi,orangutan,otter,owl,ox,oyster,panda,parrot,patas_monkey,pelecaniformes,penguin,pig,pigeon,porcupine,possum,pygmy_marmoset,raccoon,rat,raven,red_fox,reindeer,rhino,ringtail,sandpiper,sea_lions,seahorse,seal,shark,sheep,silvery_marmoset,snake,sparrow,squid,squirrel,starfish,swan,teddy_bear,tiger,turkey,turtle,virginia_opossum,whale,white_headed_capuchin,wolf,wombat,woodpecker,zebra
| Object name | Description |
|---|---|
| antelope | Antelope |
| badger | Badger |
| bald_eagle | Bald Eagle |
| bald_uakari | Bald Uakari |
| bat | Bat |
| bee | Bee |
| beetle | Beetle |
| bison | Bison |
| black_bear | Black Bear |
| black_headed_night_monkey | Black Headed Night Monkey |
| boar | Boar |
| bobcat | Bobcat |
| buffalo | Buffalo |
| butterfly | Butterfly |
| canada_lynx | Canada Lynx |
| cat | Cat |
| caterpillar | Caterpillar |
| chimpanzee | Chimpanzee |
| cockroach | Cockroach |
| columbian_black_tailed_deer | Columbian Black Tailed Deer |
| common_squirrel_monkey | Common Squirrel Monkey |
| cougar | Cougar |
| cow | Cow |
| coyote | Coyote |
| crab | Crab |
| crow | Crow |
| deer | Deer |
| dog | Dog |
| dolphin | Dolphin |
| donkey | Donkey |
| dragonfly | Dragonfly |
| duck | Duck |
| elephant | Elephant |
| elk | Elk |
| flamingo | Flamingo |
| fly | Fly |
| goat | Goat |
| goldfish | Goldfish |
| goose | Goose |
| gorilla | Gorilla |
| grasshopper | Grasshopper |
| gray_fox | Gray Fox |
| gray_wolf | Gray Wolf |
| grizzly_ bear | Grizzly Bear |
| hamster | Hamster |
| hare | Hare |
| hedgehog | Hedgehog |
| hippopotamus | Hippopotamus |
| hornbill | Hornbill |
| horse | Horse |
| hummingbird | Hummingbird |
| hyena | Hyena |
| japanese_macaque | Japanese Macaque |
| jellyfish | Jellyfish |
| kangaroo | Kangaroo |
| koala | Koala |
| ladybugs | Ladybugs |
| leopard | Leopard |
| lion | Lion |
| lizard | Lizard |
| lobster | Lobster |
| mantled_howler | Mantled Howler |
| mosquito | Mosquito |
| moth | Moth |
| mountain_beaver | Mountain Beaver |
| mouse | Mouse |
| nilgiri_langur | Nilgiri Langur |
| nutria | Nutria |
| octopus | Octopus |
| okapi | Okapi |
| orangutan | Orangutan |
| otter | Otter |
| owl | Owl |
| ox | Ox |
| oyster | Oyster |
| panda | Panda |
| parrot | Parrot |
| patas_monkey | Patas Monkey |
| pelecaniformes | Pelecaniformes |
| penguin | Penguin |
| pig | Pig |
| pigeon | Pigeon |
| porcupine | Porcupine |
| possum | Possum |
| pygmy_marmoset | Pygmy Marmoset |
| raccoon | Raccoon |
| rat | Rat |
| raven | Raven |
| red_fox | Red Fox |
| reindeer | Reindeer |
| rhino | Rhino |
| ringtail | Ringtail |
| sandpiper | Sandpiper |
| sea_lions | Sea Lions |
| seahorse | Seahorse |
| seal | Seal |
| shark | Shark |
| sheep | Sheep |
| silvery_marmoset | Silvery Marmoset |
| snake | Snake |
| sparrow | Sparrow |
| squid | Squid |
| squirrel | Squirrel |
| starfish | Starfish |
| swan | Swan |
| teddy_ bear | Teddy Bear |
| tiger | Tiger |
| turkey | Turkey |
| turtle | Turtle |
| virginia_opossum | Virginia Opossum |
| whale | Whale |
| white_headed_capuchin | White Headed Capuchin |
| wolf | Wolf |
| wombat | Wombat |
| woodpecker | Woodpecker |
| zebra | Zebra |
# crowd-v2
This is a model which used to detect heads not humans or bodies. You may find a demo pipeline A sample pipeline for people counting with heads detection could be found here (opens new window).
# Labels
The label string is as follows.
HeadBox,FullPersonBox
| Object name | Description |
|---|---|
| HeadBox | Head of a person |
| FullPersonBox | Full body of a person |
# Remarks
For HeadBox, the confidence level would be set to around 12% to get the optimal result.
# fish
This is an image classifier with object detection model for 13 different fish species including Surgeonfishes, Triggerfishes, Jacks, Spadefishes, Wrasse, Snappers, Angelfishes, Damselfishes, Parrotfishes, Tunas, Groupers, Selachimorpha and Moorish Idol. The model will identify bounded regions of interest within the input image inside of which is a fish and then classifies the types of fish in the bounding box. Around 2,000 samples were used for training in the dataset with 0.90 loss in performance.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
Acanthuridae -Surgeonfishes-,Balistidae -Triggerfishes-,Carangidae -Jacks-,Ephippidae -Spadefishes-,Labridae -Wrasse-,Lutjanidae -Snappers-,Pomacanthidae -Angelfishes-,Pomacentridae -Damselfishes-,Scaridae -Parrotfishes-,Scombridae -Tunas-,Serranidae -Groupers-,Shark -Selachimorpha-,Zanclidae -Moorish Idol-
| Object name | Description |
|---|---|
| Acanthuridae -Surgeonfishes- | Surgeonfishes |
| Balistidae -Triggerfishes- | Triggerfishes |
| Carangidae -Jacks- | Jacks |
| Ephippidae -Spadefishes- | Spadefishes |
| Labridae -Wrasse- | Wrasse |
| Lutjanidae -Snappers- | Snappers |
| Pomacanthidae -Angelfishes- | Angelfishes |
| Pomacentridae -Damselfishes- | Damselfishes |
| Scaridae -Parrotfishes- | Parrotfishes |
| Scombridae -Tunas- | Tunas |
| Serranidae -Groupers- | Groupers |
| Shark -Selachimorpha- | Selachimorpha |
| Zanclidae -Moorish Idol- | Moorish Idol |
# CheckYourAge-v3
This face recognition model predicts appearance age based on your facial characteristics which classifies into different age groups as follows: baby (0-1), toddler (2-3), young children (4-5), children (6-12), teen (13-17), young adult (18-24), adults (25-30), middle-aged adults (31-45), old adults (46-64), senior adults (65-79), aged 80 or above. More than 88,000 samples were used for training this image classifier in the dataset with an Average Precision of 79.5%.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
baby (0-1),toddler (2-3),young children (4-5),children (6-12),teen (13-17),young adult (18-24),adults (25-30),middle-aged adults (31-45),old adults (46-64),senior adults (65-79),aged 80 or above
| Object name | Description |
|---|---|
| baby (0-1) | Baby (0-1) |
| toddler (2-3) | Toddler (2-3) |
| young children (4-5) | Young Children (4-5) |
| children (6-12) | Children (6-12) |
| teen (13-17) | Teen (13-17) |
| young adult (18-24) | Young Adult (18-24) |
| adults (25-30) | Adults (25-30) |
| middle-aged adults (31-45) | Middle-aged Adults (31-45) |
| old adults (46-64) | Old Adults (46-64) |
| senior adults (65-79) | Senior Adults (65-79) |
| aged 80 or above | Aged 80 or above |
# roadsign
This road sign provides critical information on roads. With a model that could reliably detect traffic signs, it could be further implemented (for example with IFTTT integration) to alert drivers for a safer driving behaviour e.g. giving voice signal on how fast they could safely go, or informed about the traffic lights, stop sign or if there is any crosswalk ahead. More than 2,500 samples were used for training object detector in the dataset.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
crosswalk,speedlimit,stop,trafficlight
| Object name | Description |
|---|---|
| crosswalk | Crosswalk |
| speedlimit | Speed limit sign |
| stop | Stop sign |
| trafficlight | Traffic light |
# gunholding
An objection detection model to detect if anyone is holding a pistol or to detect the location of a pistol within an image or video frame. Around 3,000 samples were used for training this object detector in the dataset with an Average Precision of 78.6%.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
pistol
| Object name | Description |
|---|---|
| pistol | Pistol |
# MusicalInstruments
This is an image classifier for 29 different musical instruments including banjo, concertina, harmonica, sitar, tambourine, drums, harp, xylophone, casaba, dulcimer, marakas, trombone, acordian, castanets, flute, ocarina, trumpet, alphorn, clarinet, guiro, piano, tuba, bagpipes, clavichord, guitar, saxaphone, violin, bongo drum and steel drum. The model will classify the type of musical instrument and output the category label for that image. Over 15,000 samples were used for training in the dataset with an average precision of 89.3%.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
banjo,concertina,harmonica,sitar,Tambourine,drums,harp,Xylophone,casaba,dulcimer,marakas,trombone,acordian,castanets,flute,ocarina,trumpet,alphorn,clarinet,guiro,piano,tuba,bagpipes,clavichord,guitar,saxaphone,violin,bongo drum,steel drum
| Object name | Description |
|---|---|
| banjo | Banjo |
| concertina | Concertina |
| harmonica | Harmonica |
| sitar | Sitar |
| Tambourine | Tambourine |
| drums | Drums |
| harp | Harp |
| Xylophone | Xylophone |
| casaba | Casaba |
| dulcimer | Dulcimer |
| marakas | Marakas |
| trombone | Trombone |
| acordian | Acordian |
| castanets | Castanets |
| flute | Flute |
| ocarina | Ocarina |
| trumpet | Trumpet |
| alphorn | Alphorn |
| clarinet | Clarinet |
| guiro | Guiro |
| piano | Piano |
| tuba | Tuba |
| bagpipes | Bagpipes |
| clavichord | Clavichord |
| guitar | Guitar |
| saxaphone | Saxaphone |
| violin | Violin |
| bongo drum | Bongo Drum |
| steel drum | Steel Drum |
# HelmetDetector
An objection detection model to detect if a person is wearing a safety helmet or not. A person holding a helmet or merely a helmet image would not be detected as wearing a helmet in this model. More than 21,000 samples were used for training this object detector in the dataset with a loss of 1.23.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
helmet
| Object name | Description |
|---|---|
| helmet | Helmet |
# gloves-v2
An objection detection model to detect if a person is wearing safety gloves or not. A person holding gloves or merely a glove image would not be considered as wearing gloves in this model. Around 1,000 samples were used for training in the dataset with 0.51 loss in performance. The average precision for "wearing gloves" and "not wearing gloves" are 88% and 86% respectively.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
WearingGloves,NotWearingGloves
| Object name | Description |
|---|---|
| WearingGloves | Wearing gloves |
| NotWearingGloves | Not wearing gloves |
# goggles
An objection detection model to detect if a person is wearing safety glasses/goggles or not for industrial safety gear checking. A person holding goggles or merely an image with goggles inside would not be considered as wearing glasses/goggles in this model. Around 2,000 samples were used for training in the dataset with 0.22 loss in performance.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
safetyglasses
| Object name | Description |
|---|---|
| safetyglasses | Wearing safety glasses |
# BoarDetection
An objection detection model to detect if there is any wild pig/boar. The model will identify bounded regions of interest within the input image inside of which is an animal and then classifies if it is a boar or non-boar animal in the bounding box. Over 2,500 samples were used for training in the dataset with an Average Precision of 99.8%.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
wild_pigs,non_pigs,Negative
| Object name | Description |
|---|---|
| wild_pigs | Boar |
| non_pigs | Other non-boar animal |
| Negative | No animal detected |
# FlyingVehicles
This is an image classifier for 6 different types of flying vehciles including drone, fighter jet, helicopter, missile, passenger plane and rocket. The model will classify the type of flying vehcile and output the category label for that image. Over 8,000 samples were used for training in the dataset. The average precision and recall rate are 85.3% and 75.0% respectively.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
drone,fighter_jet,helicopter,missile,passenger_plane,rocket
| Object name | Description |
|---|---|
| drone | Drone |
| fighter_jet | Fighter jet |
| helicopter | Helicopter |
| missile | Missile |
| passenger_plane | Passenger plane |
| rocket | Rocket |
# MeatDetector
An objection detection model to detect meat, beef, or Wagyu food. The model will identify bounded regions of interest within the input image inside of which is a meat, beef, or Wagyu food. Over 1,000 samples were used for training in the dataset with 0.64 loss in performance.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
meat
| Object name | Description |
|---|---|
| meat | Meat |
# MeatQuality-v2
This is an image classifier for good or bad quality of wagyu beef, a Japanese beef. The model was trained using over 300 photos of wagyu beef labelled in good quality or bad quality meat. The average precision and recall rate are both 100%.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
good,poor
| Object name | Description |
|---|---|
| good | Good wagyu |
| poor | Poor wagyu |
# Remarks
MeatQuality-v2 is an improved version of MeatQuality model mainly for classifying good or bad quality wagyu.
# wallcracks
Cracks are one of the important criteria used for diagnosing the deterioration of concrete structures. This image classification model will automatically detect wall cracks which will enhance the efficiency of inspection work related to the maintenance and management of social infrastructure. The model was trained using over 8,900 photos of different walls labelled with cracks or no cracks. The average precision and recall rate are both 96.6%.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
cracks_detected,normal
| Object name | Description |
|---|---|
| cracks_detected | Wall cracks |
| normal | Normal wall |
# ChestXrayTestForPneumonia
This is an image classification model to check if you have pneumonia and whether it is caused by a virus (an infection that causes inflammation in one or both of the lungs) or caused by certain bacteria based on the chest X-Ray. Over 8,300 samples were used for training in the dataset. The average precision and recall rate are both over 78%.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
normal,virus_pneumonia,bacteria_pneumonia
| Object name | Description |
|---|---|
| normal | No pneumonia |
| virus_pneumonia | Virus pneumonia |
| bacteria_pneumonia | Bacteria pneumonia |
# VehicleLicensePlate
An objection detection model to detect vehicle license plates (registration number plates) of any vehicles. Around 1,200 samples were used for training in the dataset with 0.27 loss in performance and an average precision of 95.68%.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
licence
| Object name | Description |
|---|---|
| licence | Vehicle license plate |
# Remarks
VehicleLicensePlate is an improved version of licenseplate model to detect vehicle license plates.
# BrainTumorMRI
Early detection and classification of brain tumors is an important research domain in the field of medical imaging and accordingly helps in selecting the most convenient treatment method to save patients life. Brain tumors can be cancerous (malignant) or noncancerous (benign). When benign or malignant tumors grow, they can cause the pressure inside your skull to increase. This can cause brain damage, and it can be life-threatening. This image classification model was trained to automatically classify if there is any brain tumors using the Magnetic Resonance Imaging (MRI) images. The following types of brain tumor could be recognised using our BrainTumorMRI model:
- Glioma: A very common type of tumor originating in the brain. Around 33% of all brain tumors are gliomas, which is a type of tumor that occurs in the brain and spinal cord, classified according to the type of glial cell involved in the tumor.
- Meningioma: The most common type of tumor which may compress or squeeze the adjacent brain, nerves and vessels. Most meningiomas grow very slowly, often over many years without causing symptoms. That is why early detection for meningioma using this model would be very useful.
- Pituitary: An abnormal growths that develop in your pituitary gland. Not all pituitary tumors cause symptoms. That is also why it is very important for an early detection of pituitary using an imaging test such as an MRI.
Around 39,500 MRI images have been used to trained the model with an average precision and recall rate of 99.9%.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
no_tumor,glioma,meningioma,pituitary
| Object name | Description |
|---|---|
| no_tumor | No brain tumor |
| glioma | Glioma |
| meningioma | Meningioma |
| pituitary | Pituitary |
# rock
This is an image classifier for 3 different types of rock including andesite, basalt and granite. The model will classify the type of rock and output the category label for that image. Around 400 samples were used for training in the dataset with an average precision of 87.9%.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
andesite,basalt,granite
| Object name | Description |
|---|---|
| andesite | Andesite |
| basalt | Basalt |
| granite | Granite |
# FakeFingerprintDetection
This is an image classifier for real or fake fingerprint image. This model is trained to tell if a fingerprint image has been manipulated in a certain way including different levels of image alteration such as obliteration, central rotation, and z-cut. Around 24,000 samples were used for training in the dataset with an average precision and recall rate of 99.8%.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
real,fake
| Object name | Description |
|---|---|
| real | Real |
| fake | Fake |
# InsectsClassifier
This is an image classification model to classify 10 different types of insect species including beetles, butterfly, cockroach, dragonfly, grasshopper, ladybird, mosquito, spiders, termite and thrips. The model will classify the type of insect and output the category label for that image. More than 5,400 samples were used for training in the dataset with an average precision and recall rate of 96.9% and 94.1% respectively.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
Beetles,Butterfly,Cockroach,Dragonfly,Grasshopper,Ladybird,Mosquito,Spiders,Termite,Thrips
| Object name | Description |
|---|---|
| Beetles | Beetles |
| Butterfly | Butterfly |
| Cockroach | Cockroach |
| Dragonfly | Dragonfly |
| Grasshopper | Grasshopper |
| Ladybird | Ladybird |
| Mosquito | Mosquito |
| Spiders | Spiders |
| Termite | Termite |
| Thrips | Thrips |
# CigaretteSmoker
This is an image classification model for detecting if there is any cigarette smoker within the input image or video frame. The model could be applied in non-smoking areas at restaurant, bar, workspace and especially in areas where a fire hazard exists in order to reduce the chance of an accidental fire and to protect diners and workers from the dangers of second-hand smoke. More than 2,900 samples were used for training in the dataset with an average precision and recall rate of 90.2% and 87.8% respectively.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
smoking,non-smoking
| Object name | Description |
|---|---|
| smoking | Smoking |
| non-smoking | Non-smoking |
# COVID-19
This is an image classification model for detecting COVID-19 using lung CT scan images to provide an accurate, fast, and easy screening and testing of COVID-19. More than 700 samples were used for training in the dataset with an average precision and recall rate of 94.9% and 90.0% respectively.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
covid_19_negative,covid_19_positive
| Object name | Description |
|---|---|
| covid_19_negative | COVID-19 negative |
| covid_19_positive | COVID-19 positive |
# BloodCellDetector
An objection detection model to detect the location of red blood cells, white blood cells and/or platelets. Around 1,000 samples were used for training in the dataset with an average precision of 87.2%.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
RBC,WBC,Platelets
| Object name | Description |
|---|---|
| RBC | Red blood cell |
| WBC | White blood cell |
| Platelets | Platelet in blood |
# WheelchairUser
This is an image classification model for recognising if there is any wheelchair user (i.e. a person who is sitting on a wheelchair) within the input image or video frame. Around 1,650 image samples were used for training in the dataset with an average precision and recall rate of 100%.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
wheelchair
| Object name | Description |
|---|---|
| wheelchair | Wheelchair |
# CoffeeFruitsRipeness
This is an image classification model for checking the ripeness of coffee fruits. More than 700 samples were used for training in the dataset with an average precision and recall rate of 99.9%. This model would be useful for effectively sorting and picking ripe coffee fruits and served as a quality check in harvesting.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
HalfRipe,Ripe,Unripe
| Object name | Description |
|---|---|
| HalfRipe | Half Ripe |
| Ripe | Ripe |
| Unripe | Unripe |
# TireConditions
This is an image classification model for recognising if there is any cracks in tire within the input image or video frame. Around 1,000 image samples were used for training in the dataset with an average precision and recall rate of 97.3% and 94.7% respectively.
Cracking in tires can be a common sign that the rubber is starting to break down or a sign of potential issue that drivers need to take immediately. The cracking might due to exposure to UV lights, oils, chemicals, and other elements that slowly break down compounds and reduce the rubber's flexibility over time. The structural integrity of the tire might already be compromised meaning it is unsafe to drive on the road. Even if there are only a small number of cracks, it could quickly lead to several major cracks that put your tire at serious risk of a sidewall blowout. Using this pre-built image classification model, you can quickly check a massive number of tires conditions regularly and see if there is any one of them need to be further checked and inspected.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
cracked,normal
| Object name | Description |
|---|---|
| cracked | Cracked tire |
| normal | Normal tire |
# BakedBeans
This is an object detection model to detect canned baked beans. The model will identify bounded regions of interest within the input image inside of which is a canned beans. Over 2,000 samples were used for training in the dataset with a total loss of 0.84 and mAP of 99.1%.
You may find a demo pipeline here (opens new window).
# Labels
The label string is as follows.
branston-baked-beans,branston-baked-beans-x4,heinz-beanz,heinz-beanz-x4,morrisons-baked-beans,morrisons-baked-beans-x4
| Object name | Description |
|---|---|
| branston-baked-beans | Branston Baked Beans (Single Can) |
| branston-baked-beans-x4 | Branston Baked Beans (Pack of 4) |
| heinz-beanz | Heinz Beanz (Single Can) |
| heinz-beanz-x4 | Heinz Beanz (Pack of 4) |
| morrisons-baked-beans | Morrisons Baked Beans (Single Can) |
| morrisons-baked-beans-x4 | Morrisons Baked Beans (Pack of 4) |
# Contact us
Should you have any inquiries regarding the pre-built models, do not hesitate to contact us via the Discord channel here (opens new window), we would love to hear from you.
← Track