Processing times
Information
provided by the participants
|
Run |
Processing times |
Computers characteristics |
|
Run 1 |
Mix of other runs… |
Pentium4,
2.8GHz, 2Go, Linux |
|
Run 2 |
Features extraction: 7.03 s /
image prediction: 13696 s i.e 1369.6
s / concept |
Pentium4,
2.8GHz, 2Go, Linux |
|
Run 3 |
Features
extraction : 6,02 s / image Learning :
~ 10 min / objet Retrieval
333,02 s / query |
Pentium4,
3.6GHz, 1Go, Linux |
|
Run 4 |
low-level
features extraction : 4.5 sec / picture model
learning : 10 sec object
detection : 0.02 sec /
picture |
Pentium4,
2.8GHz, 2Go, Linux |
|
Run 5 |
low-level features extraction
: 10 sec / picture model learning : 180 sec object detection : 1 sec / picture |
Pentium4,
2.8GHz, 2Go, Linux |
|
Run 6 |
low-level features extraction
: 10 sec / picture model learning
: 2000 sec object detection : 1 sec / picture |
Pentium4,
2.8GHz, 2Go, Linux |
|
Run 7 |
about
15min for each class (1) pre-processing: 1 or 2s / image (2) Features (color) extraction :
1s / image (3) Features (texture) extraction : 3/4s / image (4) Merging
attributes : ~ 1h i.e 5s / image. Learning : ~ 1h Retrieval : max 1 min /
query |
Xeon,
3GHz, 2Go, Linux |
|
Run 8 |
Features extraction : 2 s /
image Model learning : 200 s Retrieval : 0,05 s / image |
Pentium4,
2.4 GHz, 512Mo, Windows |
|
Run 9 |
about
1h for each class (1) pre-processing: 1 or 2s / image. (2) Features (color) extraction :
1s / image (3) Features (texture) extraction : 3/4s / image (4) Merging
attributes : ~ 1h i.e 5s / image. Learning : ~2 h (1+1) Retrieval : max 5min par
requete. |
Xeon,
3GHz, 2Go, Linux |
MAP + Retrieval time (second/image) log10 scale
MAP + Features extractions time (second/image)
MAP + Model learning time (second/object) log10 scale
