1. Experience on Crowd-Human Challenge
Zheng Ge1,2 Xin Huang1,2 Zequn Jie1,* Yuhu Shan1
1 Tencent AI Lab Waseda University
* Team leader

2. Baseline Re-implementation
Exploring Techniques for Further Improvement

3. Baseline Re-implementation
Baseline (visible body, paper): Faster-RCNN, Res50, FPN
mMR: 55.94%
mMR: 59.67%
Baseline (visible body, ours): Faster-RCNN, Res50, FPN, ROI_align
FPN+BN,
Avoiding negative anchors in ignored regions
mMR: 55.76%

4. Baseline Re-implementation
ignore region
anchor
During training RPN, we avoid the negative anchors whose IoA (Intersection over Anchor) with an arbitrary labeled ignored region > 0.5.
An example of labeled ignored region.

5. Baseline Re-implementation
mMR
Baseline (paper)
55.94
85.60
Baseline (ours)
59.67
83.70
+ Avoid negative anchors
58.53
84.92
+ BN FPN
55.76
85.43
Table1. Evaluation results on Crowdhuman (visible body) validation set.
Baseline (ours): Faster RCNN, Res50, FPN, ROI_align, nms_pre=6000

6. Baseline Re-implementation
The effect of nms_pre.
nms_pre
mMR
12000
56.45
85.37
6000
55.76
85.43
2000
54.24
84.00
1500
54.00
83.38
1000
53.86
82.03
The best result on previous page
Table2. Evaluation results on Crowdhuman (visible body) validation set.

7. Baseline Re-implementation
Conclusion
mMR
Baseline (paper)
55.94
85.60
Baseline (ours new)
54.24
84.00
Table3. Baseline comparison on validation set (visible body).
mMR
Baseline (paper)
50.42
84.95
Baseline (ours new)
46.52
84.04
Table4. Baseline comparison on validation set (full body).
Baseline (ours new): Faster RCNN, Res50, BN FPN, ROI_align, nms_pre=2000, Avoid negative anchors

8. Baseline Re-implementation
Exploring Techniques for Further Improvement

9. Exploring Techniques Techniques bringing Techniques bringing
large improvements
Techniques bringing
marginal improvements
Techniques of limited effects
Cascade R-CNN
Deformable Conv Net
SENet154
Multi-Scale Train/Test
Ensemble
SyncBN
Focal Loss
GIoU/IoU Loss
COCO Pretrain
Soft-NMS
OHEM
Adaptive-NMS
Guided Anchor RPN
Scale Balanced Sampling …
Bounded Repulsion Loss
R-CNN Context
Merge Cityperson dataset

10. Exploring Techniques Conclusion mMR gain Baseline 46.52 -
+ Bounded RepGT Loss
45.80
0.72
+ R-CNN Context
45.45
0.35
+ multi-scale train/test
43.32
2.13
+ Cascade R-CNN
+ Deformable Conv
+ SENet154
37.17
6.15
mMR
gain
Baseline
46.52 -
+ CityPerson
45.58
0.94
Baseline :
Faster RCNN, Res50, BN FPN, ROI_align,
nms_pre=2000, Avoid anchors

11. Exploring Techniques Final Ensemble for Submission
SENet154 + Cascade R-CNN + DCN
SENet154 + Cascade R-CNN + DCN + Bounded RepGt + Context
SENet154 + Cascade R-CNN + DCN + Bounded RepGt + Context + cocopretrain
SENet154 + Cascade R-CNN + DCN + Bounded RepGt + Context + cityperson
SEResNeXt101 + Cascade R-CNN + DCN + Bounded RepGt + Context
Coco没有增益，我们依旧使用它做ensemble的原因是希望coco能提供diversity, 使用senext101的理由也是如此

12. Exploring Techniques About Jaccard Index Score Threshold Thr. JI 0.55
75.78
0.50
76.48
0.45
77.01
0.40
77.46
Val.
Set
Val. Subset
Test
Set
filtering
huge gap
best threshold:
0.55

13. Thanks