PASCAL-Context Dataset
This dataset is a set of additional annotations for PASCAL VOC 2010. It goes beyond the original PASCAL semantic segmentation task by providing annotations for the whole scene. The statistics section has a full list of 400+ labels.
Below are some example segmentations from the dataset.
Below are some example class masks.
computer
sofa
treeDownload
Training and Validation Set
The following download contains only the annotations for the training/validation set and a text file (labels.txt) containing the integer-to-name mapping.
- MD5:
df034edb2c12aa7d33b42b20bb1796e3 - SHA1:
e1375777199dcc02097e68aea755415ca173aa2d
Original images for the dataset can be downloaded from the PASCAL VOC 2010 website.
Testing Set
Please download the images from the PASCAL VOC website.
Project Specific Downloads
The segmentation results for the 33 categories and the label-to-name mapping can be downloaded here (trained on the train subset and tested on the val subset). This is the result of a 33-way classifier. These segmentations were used to train contextDPM in the paper.
33_context_labels.tar.gz (27.1 MB)
The segmentation results for the 59 categories and the label-to-name mapping can be downloaded here (trained on the train subset and tested on the val subset). This is the result of a 59-way classifier. The accuracies in Tables 1 and 2 in the CVPR paper are computed according to these predictions.
Statistics
Since the dataset is an annotation of PASCAL VOC 2010, it has the same statistics as those of the original dataset. Training and validation contains 10,103 images while testing contains 9,637 images.
Click on the panel below to expand the full class list. The following image count and average area are calculated only over the training and validation set. When the testing set is released these numbers will be updated. Image counts below may be zero because a class was present in the testing set but not the training and validation set.
| Class Name | # of Images | Average Area (% of image) |
|---|---|---|
empty | 0 | |
accordion | 2 | 5.26% |
aeroplane | 597 | 17.41% |
air conditioner | 5 | 2.68% |
antenna | 2 | 0.80% |
artillery | 1 | 3.36% |
ashtray | 18 | 1.11% |
atrium | 1 | 13.64% |
baby carriage | 13 | 12.28% |
bag | 319 | 3.18% |
ball | 25 | 2.12% |
balloon | 18 | 2.88% |
bamboo weaving | 1 | 12.78% |
barrel | 3 | 3.58% |
baseball bat | 1 | 1.05% |
basket | 89 | 5.70% |
basketball backboard | 6 | 5.37% |
bathtub | 5 | 36.08% |
bed | 120 | 28.08% |
bedclothes | 296 | 27.01% |
beer | 1 | 2.41% |
bell | 2 | 1.24% |
bench | 119 | 5.49% |
bicycle | 503 | 17.20% |
binoculars | 0 | |
bird | 664 | 12.36% |
bird cage | 21 | 44.05% |
bird feeder | 14 | 18.47% |
bird nest | 6 | 21.69% |
blackboard | 2 | 12.70% |
board | 137 | 9.67% |
boat | 466 | 16.67% |
bone | 3 | 2.55% |
book | 416 | 3.80% |
bottle | 821 | 7.35% |
bottle opener | 1 | 0.61% |
bowl | 137 | 2.57% |
box | 434 | 5.16% |
bracelet | 1 | 0.76% |
brick | 9 | 4.38% |
bridge | 31 | 10.39% |
broom | 5 | 6.75% |
brush | 120 | 10.28% |
bucket | 175 | 2.73% |
building | 3064 | 20.22% |
bus | 399 | 32.86% |
cabinet | 536 | 13.62% |
cabinet door | 1 | 5.84% |
cage | 29 | 32.76% |
cake | 2 | 8.76% |
calculator | 1 | 1.30% |
calendar | 15 | 3.30% |
camel | 0 | |
camera | 52 | 1.80% |
camera lens | 0 | |
can | 78 | 2.24% |
candle | 43 | 1.04% |
candle holder | 29 | 0.93% |
cap | 16 | 3.45% |
car | 1273 | 19.45% |
card | 11 | 3.41% |
cart | 11 | 6.56% |
case | 91 | 4.12% |
casette recorder | 1 | 1.82% |
cash register | 1 | 0.88% |
cat | 1006 | 33.30% |
cd | 45 | 3.33% |
cd player | 1 | 1.35% |
ceiling | 586 | 10.00% |
cell phone | 43 | 0.49% |
cello | 4 | 15.31% |
chain | 10 | 1.52% |
chair | 1357 | 11.87% |
chessboard | 2 | 1.37% |
chicken | 0 | |
chopstick | 8 | 1.12% |
clip | 1 | 0.47% |
clippers | 2 | 9.44% |
clock | 61 | 0.99% |
closet | 1 | 27.69% |
cloth | 877 | 8.96% |
clothes tree | 0 | |
coffee | 4 | 0.59% |
coffee machine | 2 | 1.41% |
comb | 3 | 0.80% |
computer | 118 | 8.21% |
concrete | 9 | 6.79% |
cone | 30 | 1.32% |
container | 61 | 4.34% |
control booth | 0 | |
controller | 3 | 4.67% |
cooker | 0 | |
copying machine | 0 | |
coral | 0 | |
cork | 0 | |
corkscrew | 0 | |
counter | 31 | 11.19% |
court | 2 | 14.94% |
cow | 256 | 22.98% |
crabstick | 1 | 2.44% |
crane | 12 | 0.56% |
crate | 2 | 0.74% |
cross | 3 | 0.23% |
crutch | 2 | 0.52% |
cup | 432 | 2.46% |
curtain | 381 | 11.51% |
cushion | 75 | 12.60% |
cutting board | 3 | 2.50% |
dais | 1 | 10.10% |
disc | 2 | 2.49% |
disc case | 16 | 4.96% |
dishwasher | 0 | |
dock | 3 | 10.26% |
dog | 1204 | 25.23% |
dolphin | 0 | |
door | 455 | 10.01% |
drainer | 1 | 0.09% |
dray | 1 | 10.38% |
drink dispenser | 1 | 1.92% |
drinking machine | 1 | 7.36% |
drop | 0 | |
drug | 1 | 1.52% |
drum | 9 | 8.77% |
drum kit | 8 | 7.93% |
duck | 15 | 8.21% |
dumbbell | 0 | |
earphone | 8 | 1.21% |
earrings | 1 | 1.07% |
egg | 3 | 6.92% |
electric fan | 11 | 2.57% |
electric iron | 2 | 2.07% |
electric pot | 1 | 2.02% |
electric saw | 1 | 3.29% |
electronic keyboard | 0 | |
engine | 2 | 3.94% |
envelope | 0 | |
equipment | 27 | 7.52% |
escalator | 0 | |
exhibition booth | 2 | 17.59% |
extinguisher | 6 | 0.95% |
eyeglass | 15 | 2.72% |
fan | 40 | 2.05% |
faucet | 4 | 2.27% |
fax machine | 1 | 2.58% |
fence | 1021 | 10.79% |
ferris wheel | 1 | 10.09% |
fire extinguisher | 2 | 0.75% |
fire hydrant | 2 | 4.74% |
fire place | 41 | 9.53% |
fish | 7 | 4.60% |
fish tank | 3 | 25.04% |
fishbowl | 1 | 18.00% |
fishing net | 1 | 7.74% |
fishing pole | 1 | 0.21% |
flag | 100 | 2.05% |
flagstaff | 3 | 1.58% |
flame | 0 | |
flashlight | 2 | 0.38% |
floor | 1895 | 16.07% |
flower | 190 | 7.67% |
fly | 1 | 0.43% |
foam | 0 | |
food | 285 | 5.16% |
footbridge | 1 | 10.82% |
forceps | 0 | |
fork | 104 | 0.63% |
forklift | 0 | |
fountain | 1 | 1.40% |
fox | 0 | |
frame | 17 | 5.86% |
fridge | 69 | 10.07% |
frog | 1 | 0.70% |
fruit | 2 | 0.77% |
funnel | 1 | 0.37% |
furnace | 2 | 31.60% |
game controller | 4 | 0.66% |
game machine | 3 | 13.31% |
gas cylinder | 0 | |
gas hood | 0 | |
gas stove | 0 | |
gift box | 2 | 10.96% |
glass | 21 | 5.24% |
glass marble | 1 | 8.14% |
globe | 0 | |
glove | 6 | 4.34% |
goal | 4 | 10.04% |
grandstand | 30 | 18.55% |
grass | 2742 | 23.28% |
gravestone | 0 | |
ground | 2987 | 20.98% |
guardrail | 35 | 6.09% |
guitar | 53 | 4.83% |
gun | 2 | 1.07% |
hammer | 1 | 0.08% |
hand cart | 11 | 4.87% |
handle | 2 | 4.16% |
handrail | 7 | 3.75% |
hanger | 3 | 0.66% |
hard disk drive | 0 | |
hat | 27 | 4.07% |
hay | 3 | 21.08% |
headphone | 7 | 0.49% |
heater | 4 | 8.10% |
helicopter | 3 | 12.77% |
helmet | 62 | 2.32% |
holder | 2 | 2.23% |
hook | 2 | 0.61% |
horse | 429 | 22.31% |
horse-drawn carriage | 25 | 12.70% |
hot-air balloon | 1 | 84.05% |
hydrovalve | 0 | |
ice | 6 | 31.35% |
inflator pump | 0 | |
ipod | 15 | 0.75% |
iron | 2 | 2.15% |
ironing board | 1 | 9.63% |
jar | 18 | 2.80% |
kart | 1 | 16.90% |
kettle | 1 | 1.96% |
key | 4 | 1.86% |
keyboard | 140 | 4.69% |
kitchen range | 12 | 7.54% |
kite | 1 | 0.26% |
knife | 92 | 0.61% |
knife block | 2 | 0.30% |
ladder | 12 | 3.91% |
ladder truck | 1 | 0.84% |
ladle | 1 | 2.83% |
laptop | 75 | 10.55% |
leaves | 7 | 33.39% |
lid | 1 | 1.37% |
life buoy | 2 | 14.83% |
light | 908 | 1.18% |
light bulb | 1 | 0.68% |
lighter | 5 | 0.21% |
line | 1 | 0.14% |
lion | 0 | |
lobster | 0 | |
lock | 1 | 0.15% |
machine | 0 | |
mailbox | 3 | 0.39% |
mannequin | 1 | 11.90% |
map | 3 | 32.42% |
mask | 0 | |
mat | 35 | 11.11% |
match book | 1 | 2.29% |
mattress | 3 | 1.72% |
menu | 24 | 4.34% |
metal | 216 | 4.28% |
meter box | 1 | 3.43% |
microphone | 44 | 0.71% |
microwave | 26 | 3.61% |
mirror | 81 | 5.60% |
missile | 1 | 5.71% |
model | 1 | 27.81% |
money | 1 | 6.08% |
monkey | 1 | 2.75% |
mop | 3 | 0.66% |
motorbike | 471 | 24.63% |
mountain | 728 | 10.75% |
mouse | 122 | 0.43% |
mouse pad | 50 | 3.58% |
musical instrument | 7 | 11.16% |
napkin | 120 | 2.60% |
net | 3 | 6.81% |
newspaper | 30 | 3.99% |
oar | 10 | 5.48% |
ornament | 2 | 0.79% |
outlet | 26 | 0.48% |
oven | 8 | 10.24% |
oxygen bottle | 1 | 4.21% |
pack | 287 | 3.18% |
pan | 10 | 3.51% |
paper | 298 | 4.02% |
paper box | 4 | 0.99% |
paper cutter | 6 | 6.75% |
parachute | 4 | 3.45% |
parasol | 17 | 4.14% |
parterre | 1 | 14.77% |
patio | 1 | 23.16% |
pelage | 0 | |
pen | 44 | 0.33% |
pen container | 8 | 1.14% |
pencil | 2 | 0.65% |
person | 3919 | 20.64% |
photo | 3 | 2.69% |
piano | 4 | 8.83% |
picture | 674 | 5.70% |
pig | 1 | 1.41% |
pillar | 8 | 3.16% |
pillow | 55 | 11.18% |
pipe | 16 | 2.01% |
pitcher | 3 | 3.36% |
plant | 106 | 14.52% |
plastic | 11 | 6.37% |
plate | 346 | 4.44% |
platform | 154 | 15.39% |
player | 24 | 2.37% |
playground | 1 | 5.70% |
pliers | 1 | 0.39% |
plume | 1 | 1.76% |
poker | 1 | 0.73% |
poker chip | 1 | 7.81% |
pole | 1555 | 1.53% |
pool table | 1 | 29.81% |
postcard | 1 | 1.72% |
poster | 185 | 6.82% |
pot | 310 | 3.08% |
pottedplant | 625 | 11.89% |
printer | 34 | 2.92% |
projector | 2 | 0.52% |
pumpkin | 2 | 7.72% |
rabbit | 1 | 0.14% |
racket | 1 | 4.75% |
radiator | 27 | 3.93% |
radio | 6 | 2.22% |
rail | 31 | 6.12% |
rake | 1 | 0.26% |
ramp | 2 | 3.15% |
range hood | 8 | 3.98% |
receiver | 13 | 1.12% |
recorder | 2 | 23.21% |
recreational machines | 1 | 61.90% |
remote control | 67 | 0.93% |
road | 1102 | 21.97% |
robot | 0 | |
rock | 396 | 10.72% |
rocket | 0 | |
rocking horse | 1 | 0.74% |
rope | 507 | 1.35% |
rug | 181 | 20.69% |
ruler | 1 | 1.61% |
runway | 2 | 11.72% |
saddle | 85 | 1.95% |
sand | 46 | 36.43% |
saw | 1 | 1.53% |
scale | 2 | 3.64% |
scanner | 1 | 1.43% |
scissors | 5 | 0.72% |
scoop | 1 | 0.04% |
screen | 4 | 13.73% |
screwdriver | 2 | 0.32% |
sculpture | 73 | 7.67% |
scythe | 0 | |
sewer | 2 | 26.70% |
sewing machine | 3 | 1.06% |
shed | 2 | 2.25% |
sheep | 297 | 21.95% |
shell | 3 | 2.43% |
shelves | 448 | 7.64% |
shoe | 74 | 2.68% |
shopping cart | 2 | 4.54% |
shovel | 2 | 1.07% |
sidecar | 2 | 15.60% |
sidewalk | 382 | 8.65% |
sign | 592 | 2.32% |
signal light | 43 | 0.98% |
sink | 24 | 15.51% |
skateboard | 1 | 0.81% |
ski | 3 | 5.71% |
sky | 3231 | 26.77% |
sled | 2 | 4.28% |
slippers | 1 | 4.63% |
smoke | 56 | 10.42% |
snail | 0 | |
snake | 3 | 15.25% |
snow | 173 | 28.71% |
snowmobiles | 1 | 16.52% |
sofa | 831 | 25.31% |
spanner | 0 | |
spatula | 1 | 1.36% |
speaker | 118 | 3.48% |
speed bump | 0 | |
spice container | 23 | 0.99% |
spoon | 75 | 0.51% |
sprayer | 0 | |
squirrel | 1 | 6.33% |
stage | 16 | 49.30% |
stair | 36 | 21.55% |
stapler | 3 | 0.79% |
stick | 4 | 0.28% |
sticky note | 2 | 1.64% |
stone | 0 | |
stool | 9 | 3.18% |
stove | 10 | 2.22% |
straw | 6 | 0.47% |
stretcher | 0 | |
sun | 1 | 3.99% |
sunglass | 4 | 0.86% |
sunshade | 3 | 5.83% |
surveillance camera | 1 | 0.16% |
swan | 1 | 0.22% |
sweeper | 0 | |
swim ring | 2 | 9.90% |
swimming pool | 1 | 3.01% |
swing | 1 | 2.09% |
switch | 5 | 0.51% |
table | 1566 | 9.76% |
tableware | 3 | 4.19% |
tank | 2 | 2.63% |
tap | 12 | 2.30% |
tape | 9 | 0.51% |
tarp | 2 | 43.12% |
telephone | 71 | 1.72% |
telephone booth | 0 | |
tent | 36 | 12.26% |
tire | 35 | 3.29% |
toaster | 5 | 1.70% |
toilet | 2 | 12.68% |
tong | 1 | 1.80% |
tool | 3 | 3.06% |
toothbrush | 1 | 0.80% |
towel | 3 | 18.69% |
toy | 356 | 5.21% |
toy car | 2 | 30.91% |
track | 356 | 9.15% |
train | 465 | 28.46% |
trampoline | 1 | 15.35% |
trash bin | 91 | 3.03% |
tray | 9 | 5.46% |
tree | 3232 | 18.02% |
tricycle | 0 | |
tripod | 2 | 8.27% |
trophy | 2 | 1.04% |
truck | 126 | 6.39% |
tube | 3 | 3.82% |
turtle | 1 | 0.98% |
tvmonitor | 536 | 15.32% |
tweezers | 0 | |
typewriter | 1 | 3.40% |
umbrella | 13 | 4.32% |
unknown | 3734 | 8.99% |
vacuum cleaner | 6 | 4.15% |
vending machine | 2 | 23.66% |
video camera | 11 | 1.94% |
video game console | 11 | 1.77% |
video player | 37 | 2.97% |
video tape | 3 | 4.09% |
violin | 4 | 4.37% |
wakeboard | 0 | |
wall | 2794 | 22.23% |
wallet | 7 | 1.26% |
wardrobe | 2 | 18.93% |
washing machine | 17 | 15.58% |
watch | 6 | 0.81% |
water | 899 | 32.68% |
water dispenser | 5 | 3.52% |
water pipe | 3 | 2.94% |
water skate board | 1 | 0.60% |
watermelon | 2 | 7.00% |
whale | 0 | |
wharf | 0 | |
wheel | 8 | 17.21% |
wheelchair | 5 | 3.78% |
window | 756 | 11.07% |
window blinds | 1 | 1.57% |
wineglass | 197 | 2.95% |
wire | 113 | 1.17% |
wood | 462 | 8.29% |
wool | 2 | 4.78% |
Usage Considerations
The classes are not drawn from a fixed pool. Instead labelers were free to either select or type in what they believe to be the appropriate class and to determine what the appropriate object granularity is. We decided to merge/split some of the categories so the current number of categories is different from what we mentioned in the CVPR 2014 paper.
When using this dataset it is important that you examine classes to ensure they match your intended use. For example, sand is often labeled independently despite also being considered ground. Those interested in ground may want to cluster sand and ground together along with other classes.
Citation
-
PDF (2.4 MB)
Bibtex
"The Role of Context for Object Detection and Semantic Segmentation in the Wild"
Roozbeh Mottaghi, Xianjie Chen, Xiaobai Liu, Nam-Gyu Cho, Seong-Whan Lee, Sanja Fidler, Raquel Urtasun, Alan Yuille
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2014
Acknowledgements
We thank Viet Nguyen for coordinating and leading the efforts for cleaning up the annotations. We would like to acknowledge the support by Implementation of Technologies for Identification, Behavior, and Location of Human based on Sensor Network Fusion Program through the Korean Ministry of Trade, Industry and Energy (Grant Number: 10041629). We would also like to thank National Science Foundation for grant 1317376 (Visual Cortex on Silicon. NSF Expedition in Computing).