Large Scale Visual Recognition Challenge (ILSVRC) 2013: Detection spotlights
Analysis of Large Scale Visual Recognition
description
Transcript of Analysis of Large Scale Visual Recognition
![Page 1: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/1.jpg)
Analysis of Large Scale Visual Recognition
Fei-Fei Li and Olga Russakovsky
Olga Russakovsky, Jia Deng, Zhiheng Huang, Alex Berg, Li Fei-FeiDetecting avocados to zucchinis: what have we done, and where are we going? ICCV 2013 http://image-net.org/challenges/LSVRC/2012/analysis
![Page 2: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/2.jpg)
Backpack
![Page 3: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/3.jpg)
Backpack
Flute Strawberry Traffic light
Bathing capMatchstick
Racket
Sea lion
![Page 4: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/4.jpg)
Large-scale recognition
![Page 5: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/5.jpg)
Large-scale recognition
Need benchmark datasets
![Page 6: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/6.jpg)
PASCAL VOC 2005-2012
Classification: person, motorcycleDetection Segmentation
Person
Motorcycle
Action: riding bicycle
Everingham, Van Gool, Williams, Winn and Zisserman.The PASCAL Visual Object Classes (VOC) Challenge. IJCV 2010.
20 object classes 22,591 images
![Page 7: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/7.jpg)
Large Scale Visual Recognition Challenge (ILSVRC) 2010-2012
20 object classes 22,591 images1000 object classes 1,431,167 images
Dalmatian
http://image-net.org/challenges/LSVRC/{2010,2011,2012}
![Page 8: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/8.jpg)
Variety of object classes in ILSVRC
![Page 9: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/9.jpg)
Variety of object classes in ILSVRC
![Page 10: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/10.jpg)
ILSVRC Task 1: ClassificationSteel drum
![Page 11: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/11.jpg)
ILSVRC Task 1: Classification
Output:Scale
T-shirtSteel drumDrumstickMud turtle
Steel drum
✔ ✗Output:
ScaleT-shirt
Giant pandaDrumstickMud turtle
![Page 12: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/12.jpg)
ILSVRC Task 1: Classification
Output:Scale
T-shirtSteel drumDrumstickMud turtle
Steel drum
✔ ✗
Accuracy =
Output:Scale
T-shirtGiant pandaDrumstickMud turtle
Σ100,000images
1[correct on image i]1100,000
![Page 13: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/13.jpg)
ILSVRC Task 1: Classification
Accuracy (5 predictions/image)
# Su
bmiss
ions
0.72
0.74
0.85
2010
2011
2012
![Page 14: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/14.jpg)
ILSVRC Task 2: Classification + Localization
Steel drum
![Page 15: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/15.jpg)
✔ Folding chair
Persian cat
Loud speaker
Steel drumPicket
fence
OutputSteel drum
ILSVRC Task 2: Classification + Localization
![Page 16: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/16.jpg)
✔ Folding chair
Persian cat
Loud speaker
Steel drumPicket
fence
Output
✗ Folding chair
Persian cat
Loud speaker
Steel drumPicket
fence
Output (bad localization)
✗ Folding chair
Persian cat
Loud speaker
Picket fence
King penguin
Output (bad classification)
Steel drum
ILSVRC Task 2: Classification + Localization
![Page 17: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/17.jpg)
✔ Folding chair
Persian cat
Loud speaker
Steel drumPicket
fence
OutputSteel drum
ILSVRC Task 2: Classification + Localization
Accuracy = Σ100,000images
1[correct on image i]1100,000
![Page 18: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/18.jpg)
ILSVRC Task 2: Classification + Localization
ISI
OXFORD_VGG
SuperVision
Accu
racy
(5
pre
dicti
ons)
![Page 19: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/19.jpg)
What happens under the hood?
![Page 20: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/20.jpg)
What happens under the hoodon classification+localization?
![Page 21: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/21.jpg)
What happens under the hoodon classification+localization?
Olga Russakovsky, Jia Deng, Zhiheng Huang, Alex Berg, Li Fei-FeiDetecting avocados to zucchinis: what have we done, and where are we going? ICCV 2013 http://image-net.org/challenges/LSVRC/2012/analysis
![Page 22: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/22.jpg)
What happens under the hoodon classification+localization?
Preliminaries:• ILSVRC-500 (2012) dataset• Leading algorithms
Olga Russakovsky, Jia Deng, Zhiheng Huang, Alex Berg, Li Fei-FeiDetecting avocados to zucchinis: what have we done, and where are we going? ICCV 2013 http://image-net.org/challenges/LSVRC/2012/analysis
![Page 23: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/23.jpg)
What happens under the hoodon classification+localization?
Preliminaries:• ILSVRC-500 (2012) dataset• Leading algorithms
• A closer look at small objects• A closer look at textured objects
Olga Russakovsky, Jia Deng, Zhiheng Huang, Alex Berg, Li Fei-FeiDetecting avocados to zucchinis: what have we done, and where are we going? ICCV 2013 http://image-net.org/challenges/LSVRC/2012/analysis
![Page 24: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/24.jpg)
What happens under the hoodon classification+localization?
Preliminaries:• ILSVRC-500 (2012) dataset• Leading algorithms
• A closer look at small objects• A closer look at textured objects
Olga Russakovsky, Jia Deng, Zhiheng Huang, Alex Berg, Li Fei-FeiDetecting avocados to zucchinis: what have we done, and where are we going? ICCV 2013 http://image-net.org/challenges/LSVRC/2012/analysis
![Page 25: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/25.jpg)
Easy to localize Hard to localize
1000 object classes
ILSVRC (2012)
![Page 26: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/26.jpg)
Easy to localize Hard to localize
500 classes with smallest objects
ILSVRC-500 (2012)
![Page 27: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/27.jpg)
Easy to localize Hard to localize
ILSVRC-500 (2012) 500 object categories 25.3% PASCAL VOC (2012) 20 object categories 25.2%
Object scale (fraction of image area occupied by target object)
ILSVRC-500 (2012)500 classes with smallest objects
![Page 28: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/28.jpg)
Chance Performance of LocalizationSteel drum
B1 B2 B3B4 B5
B6 B7
B8 B9
N = 9 here
![Page 29: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/29.jpg)
Chance Performance of LocalizationSteel drum
B1 B2 B3B4 B5
B6 B7
B8 B9
N = 9 here
![Page 30: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/30.jpg)
Chance Performance of LocalizationSteel drum
ILSVRC-500 (2012) 500 object categories 8.4%PASCAL VOC (2012) 20 object categories 8.8%
B1 B2 B3B4 B5
B6 B7
B8 B9
N = 9 here
![Page 31: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/31.jpg)
Level of clutterSteel drum
- Generate candidate object regions using method of
Selective Search for Object Detection
vanDeSande et al. ICCV 2011- Filter out regions inside object- Count regions
![Page 32: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/32.jpg)
Level of clutterSteel drum
- Generate candidate object regions using method of
Selective Search for Object Detection
vanDeSande et al. ICCV 2011- Filter out regions inside object- Count regions
ILSVRC-500 (2012) 500 object categories 128 ± 35PASCAL VOC (2012) 20 object categories 130 ± 29
![Page 33: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/33.jpg)
What happens under the hoodon classification+localization?
Preliminaries:• ILSVRC-500 (2012) dataset – similar to PASCAL• Leading algorithms
• A closer look at small objects• A closer look at textured objects
Olga Russakovsky, Jia Deng, Zhiheng Huang, Alex Berg, Li Fei-FeiDetecting avocados to zucchinis: what have we done, and where are we going? ICCV 2013 http://image-net.org/challenges/LSVRC/2012/analysis
![Page 34: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/34.jpg)
SuperVision (SV)Alex Krizhevsky, Ilya Sutskever, Geoffrey Hinton (Krizhevsky NIPS12)
Image classification: Deep convolutional neural networks• 7 hidden “weight” layers, 650K neurons, 60M parameters,
630M connections • Rectified Linear Units, max pooling, dropout trick• Randomly extracted 224x224 patches for more data• Trained with SGD on two GPUs for a week, fully supervised
Localization: Regression on (x,y,w,h)
http://image-net.org/challenges/LSVRC/2012/supervision.pdf
![Page 35: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/35.jpg)
SuperVision (SV)Alex Krizhevsky, Ilya Sutskever, Geoffrey Hinton (Krizhevsky NIPS12)
Image classification: Deep convolutional neural networks• 7 hidden “weight” layers, 650K neurons, 60M parameters,
630M connections • Rectified Linear Units, max pooling, dropout trick• Randomly extracted 224x224 patches for more data• Trained with SGD on two GPUs for a week, fully supervised
Localization: Regression on (x,y,w,h)
http://image-net.org/challenges/LSVRC/2012/supervision.pdf
![Page 36: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/36.jpg)
OXFORD_VGG (VGG)Karen Simonyan, Yusuf Aytar, Andrea Vedaldi, Andrew Zisserman
Image classification: Fisher vector + linear SVM (Sanchez CVPR11)• Root-SIFT (Arandjelovic CVPR12), color statistics, augmentation
with patch location (x,y) (Sanchez PRL12)• Fisher vectors: 1024 Gaussians, 135K dimensions • No SPM, product quantization to compress• Semi-supervised learning to find additional bounding boxes• 1000 one-vs-rest SVM trained with Pegasos SGD• 135M parameters!
Localization: Deformable part-based models (Felzenszwalb PAMI10), without parts (root-only)
http://image-net.org/challenges/LSVRC/2012/oxford_vgg.pdf
![Page 37: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/37.jpg)
What happens under the hoodon classification+localization?
Preliminaries:• ILSVRC-500 (2012) dataset – similar to PASCAL• Leading algorithms: SV and VGG
• A closer look at small objects• A closer look at textured objects
Olga Russakovsky, Jia Deng, Zhiheng Huang, Alex Berg, Li Fei-FeiDetecting avocados to zucchinis: what have we done, and where are we going? ICCV 2013 http://image-net.org/challenges/LSVRC/2012/analysis
![Page 38: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/38.jpg)
SV VGG
Cls+
loc
accu
racy
54.3%45.8%
Results on ILSVRC-500
![Page 39: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/39.jpg)
Difference in accuracy: SV versus VGG
Classification-only
✔ Folding chair
Persian cat
Loud speaker
Steel drumPicket
fence
![Page 40: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/40.jpg)
Object scale
Cls.
Acc
urac
y: S
V - V
GG
Difference in accuracy: SV versus VGG
Classification-only
![Page 41: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/41.jpg)
SV better(452 classes)
VGG better(34 classes)
Object scale
Cls.
Acc
urac
y: S
V - V
GG
Difference in accuracy: SV versus VGG
Classification-only
![Page 42: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/42.jpg)
SV better(452 classes)
VGG better(34 classes)
Object scale
Cls.
Acc
urac
y: S
V - V
GG
Difference in accuracy: SV versus VGG
Classification-only
*
*** *** ***
*** *** ***
SV beats VGG
VGG beats SV
![Page 43: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/43.jpg)
SV better(452 classes)
VGG better(34 classes)
Object scale
Cls.
Acc
urac
y: S
V - V
GG
Difference in accuracy: SV versus VGG
Cls+
Loc
Accu
racy
: SV
- VGG
Object scale
Classification-only
VGG better(150 classes)
SV better(338 classes)
Classification+Localiation
![Page 44: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/44.jpg)
Cumulative accuracy across scales
SV
VGGSV
VGG
Object scale
Cum
ulati
ve c
ls. a
ccur
acy
Classification-only Classification+Localization
Cum
ulati
ve c
ls+lo
c ac
cura
cy
Object scale
![Page 45: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/45.jpg)
Cumulative accuracy across scales
SV
VGGSV
Object scale
Cum
ulati
ve c
ls. a
ccur
acy
Classification-only Classification+Localization
Cum
ulati
ve c
ls+lo
c ac
cura
cy
Object scale0.24
205 smallest object classes
VGG
![Page 46: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/46.jpg)
What happens under the hoodon classification+localization?
Preliminaries:• ILSVRC-500 (2012) dataset – similar to PASCAL• Leading algorithms: SV and VGG
• SV always great at classification, but VGG does better than SV at localizing small objects
• A closer look at textured objects
Olga Russakovsky, Jia Deng, Zhiheng Huang, Alex Berg, Li Fei-FeiDetecting avocados to zucchinis: what have we done, and where are we going? ICCV 2013 http://image-net.org/challenges/LSVRC/2012/analysis
![Page 47: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/47.jpg)
What happens under the hoodon classification+localization?
Preliminaries:• ILSVRC-500 (2012) dataset – similar to PASCAL• Leading algorithms: SV and VGG
• SV always great at classification, but VGG does better than SV at localizing small objects
• A closer look at textured objectsWHY?
Olga Russakovsky, Jia Deng, Zhiheng Huang, Alex Berg, Li Fei-FeiDetecting avocados to zucchinis: what have we done, and where are we going? ICCV 2013 http://image-net.org/challenges/LSVRC/2012/analysis
![Page 48: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/48.jpg)
What happens under the hoodon classification+localization?
Preliminaries:• ILSVRC-500 (2012) dataset – similar to PASCAL• Leading algorithms: SV and VGG
• SV always great at classification, but VGG does better than SV at localizing small objects
• A closer look at textured objects
Olga Russakovsky, Jia Deng, Zhiheng Huang, Alex Berg, Li Fei-FeiDetecting avocados to zucchinis: what have we done, and where are we going? ICCV 2013 http://image-net.org/challenges/LSVRC/2012/analysis
![Page 49: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/49.jpg)
Textured objects (ILSVRC-500)
Amount of textureLow High
![Page 50: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/50.jpg)
No texture Low texture Medium texture High texture# classes 116 189 143 52
Textured objects (ILSVRC-500)
Amount of textureLow High
![Page 51: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/51.jpg)
No texture Low texture Medium texture High texture# classes 116 189 143 52
Object scale 20.8% 23.7% 23.5% 25.0%
Textured objects (ILSVRC-500)
Amount of textureLow High
![Page 52: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/52.jpg)
No texture Low texture Medium texture High texture# classes 116 189 149 143 115 52 35
Object scale 20.8% 23.7% 20.8% 23.5% 20.8% 25.0% 20.8%
Textured objects (416 classes)
Amount of textureLow High
![Page 53: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/53.jpg)
Localizing textured objects (416 classes, same average object scale at each level of texture)
Loca
lizati
on a
ccur
acy
Level of texture
SV VGG
![Page 54: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/54.jpg)
Level of texture
Loca
lizati
on a
ccur
acy On correctly classified images
SV VGG
Localizing textured objects (416 classes, same average object scale at each level of texture)
![Page 55: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/55.jpg)
Level of texture
Loca
lizati
on a
ccur
acy On correctly classified images
SV VGG
Localizing textured objects (416 classes, same average object scale at each level of texture)
![Page 56: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/56.jpg)
What happens under the hoodon classification+localization?
Preliminaries:• ILSVRC-500 (2012) dataset – similar to PASCAL• Leading algorithms: SV and VGG
• SV always great at classification, but VGG does better than SV at localizing small objects
• Textured objects easier to localize, especially for SV
Olga Russakovsky, Jia Deng, Zhiheng Huang, Alex Berg, Li Fei-FeiDetecting avocados to zucchinis: what have we done, and where are we going? ICCV 2013 http://image-net.org/challenges/LSVRC/2012/analysis
![Page 57: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/57.jpg)
ILSVRC 2013 with large-scale object detection
http://image-net.org/challenges/LSVRC/2013/
Fully annotated 200 object classes across 60,000 images
Allows evaluation of generic object detection in cluttered scenes at scale
PersonCar
MotorcycleHelmet
NEW
![Page 58: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/58.jpg)
ILSVRC 2013 with large-scale object detection
Statistics PASCAL VOC 2012 ILSVRC 2013Object classes 20 200
TrainingImages 5.7K 395KObjects 13.6K 345K
ValidationImages 5.8K 20.1KObjects 13.8K 55.5K
TestingImages 11.0K 40.1KObjects --- ---
4x
10x
http://image-net.org/challenges/LSVRC/2013/
25x
More than 50,000 person instances annotated
NEW
![Page 59: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/59.jpg)
• 159 downloads so far:http://image-net.org/challenges/LSVRC/2013/
• Submission deadline Nov. 15th
• ICCV workshop on December 7th, 2013
• Fine-Grained Challenge 2013:https://sites.google.com/site/fgcomp2013/
ILSVRC 2013 with large-scale object detection
NEW
![Page 60: Analysis of Large Scale Visual Recognition](https://reader035.fdocuments.in/reader035/viewer/2022062301/56816156550346895dd0de7b/html5/thumbnails/60.jpg)
Thank you!
Prof. Alex BergUNC Chapel Hill
Jonathan KrauseStanford U.
Sanjeev SatheeshStanford U.
Zhiheng HuangStanford U.
Dr. Jia DengStanford U.
Hao SuStanford U.