Optimized Scalable and Learnable Binary Quantization Network for LiDAR Point Cloud

Zhi Zhao; Yanxin Ma; Ke Xu; Jianwei Wan

doi:10.3788/AOS202242.1212005

Acta Optica Sinica, Volume. 42, Issue 12, 1212005(2022)

Optimized Scalable and Learnable Binary Quantization Network for LiDAR Point Cloud

Zhi Zhao^1、*, Yanxin Ma², Ke Xu¹, and Jianwei Wan¹

Author Affiliations

¹College of Electronic Science, National University of Defense Technology, Changsha 410073, Hunan, China

²College of Meteorology and Oceanography, National University of Defense Technology, Changsha 410073, Hunan, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(26)

Fig. 1. General framework for point cloud learnable binary quantization network model

Download full size

Fig. 2. Learnable binary quantization network model of PointNet

Download full size

Fig. 3. Gene-algorithm based binary quantization scale factor recovery

Download full size

Fig. 4. Optimal search of scale factor based on gene-algorithm optimization

Download full size

Fig. 5. Curves of feature entropy before and after pooling. (a) n=5; (b) n=20; (c) n=50; (d) n=100

Download full size

Fig. 6. Minimization of statistically self-adaptive pooling loss. (a) Quantitative network self-regulation; (b) statistical knowledge transfer regulation of full precision network

Download full size

Fig. 7. Comparison of adjusted feature probability distributions. (a) Feature distribution comparison 1; (b) feature distribution comparison 2; (c) feature distribution comparison 3

Download full size

Fig. 8. Learnable training process

Download full size

Fig. 9. Comparison of optimized pooling. (a) Quantization network self-adjustment; (b) full-precision network transfer adjustment

Download full size

Fig. 10. Training performance comparison. (a) Comparison result 1; (b) comparison result 2; (c) comparison result 3

Download full size

Fig. 11. Scaling factor searching based on gene-optimized algorithm. (a) Iterative searching process; (b) feature maps produced by binary conv layer

Download full size

Fig. 12. Comparison of different channel feature maps of different binary convolution layers (sub-figures from left to right are 3 corresponding channels in sequence). (a) Feature maps of different channels of 1^stbinary convolution layer; (b) feature maps of different channels of 2^nd binary convolution layer; (c) feature maps of different channels of 3^rd binary convolution layer

Download full size

Fig. 13. Comparison of feature maps of binary convolution layers at different locations (sub-figures from left to right are 3 convolution layers in sequence). (a) Feature map of location 1; (b) feature map of location 2; (c) feature map of location 3

Download full size

Fig. 14. Feature maps of pooling. (a) Activation features before pooling; (b) pooling features of non-optimized binary network; (c) pooling features of binary network with pooling optimization; (d) pooling features of binary network with scaling and pooling optimization

Download full size

Fig. 15. Partial results of part segmentation. (a) Knife; (b) motorbike; (c) lamp

Download full size

Fig. 16. Partial results of semantic segmentation. (a) Area 1_Conference Room 2; (b) Area 1_Office Room 2; (c) Area 1_Hallway 1

Download full size

Fig. 17. Overall performance comparisons. (a) Performance comparison 1; (b) performance comparison 2

Download full size

Fig. 18. Inference time comparisons

Download full size

Table 1. Comparison of binary quantization algorithms

View table

Table 1. Comparison of binary quantization algorithms

Method	Bit width N_w /bit	Bit width N_a/bit	Scaling/Shiftingfactor	Floating pointcalculation	Bitwisecalculation
BNN	1	1	—	0	O₁×O₂
XNOR-Net	1	1	Scaling	O₁	O₁×O₂
IRNet	1	1	Shifting	0	O₁×O₂+O₁
BiPointNet	1	1	Scaling	O₁	O₁×O₂
			Pooling shifting	S	0
			Scaling	O₁	O₁×O₂
Proposed model	1	1	Pooling shifting	S	0
			Pooling scaling	S	0

Table 2. Comparison of binary quantization methods without optimized pooling

View table

Table 2. Comparison of binary quantization methods without optimized pooling

Method	Pooling type	Bit width N_w /bit	Bit width N_a /bit	Precision P_c /%
Full precision	MAX	32	32	88.2
BNN	MAX	1	1	26.8
IRNet	MAX	1	1	18.5
XNOR-Net	MAX	1	1	71.8
BiPointNet	MAX	1	1	4.1
Proposed method (Hist)	MAX	1	1	79.9
Proposed method (KDE)	MAX	1	1	80.2
Proposed method (KNN)	MAX	1	1	81.7

Table 3. Comparison of binary quantization methods with optimized pooling

View table

Table 3. Comparison of binary quantization methods with optimized pooling

Method	Pooling type	Bit width N_w /bit	Bit width N_a /bit	Precision P_c /%
Full precision	MAX	32	32	88.2
	MAX	1	1	26.8
BNN	APSS1^*	1	1	80.2
	APSS2^*	1	1	78.1
	MAX	1	1	18.5
IRNet	APSS1^*	1	1	82.3
	APSS2^*	1	1	80.7
	MAX	1	1	71.8
XNOR-Net	APSS1^*	1	1	86.0
	APSS2^*	1	1	85.6
	MAX	1	1	4.1
BiPointNet	APSS1^*	1	1	81.3
	APSS2^*	1	1	82.7

Table 4. Comparison of typical binary quantization methods

View table

Table 4. Comparison of typical binary quantization methods

Method	Pooling type	Bit width N_w /bit	Bit width N_a /bit	Precision P_c /%
Full precision	MAX	32	32	88.2
BNN	MAX	1	1	26.8
XNOR-Net	MAX	1	1	71.8
IRNet	MAX	1	1	18.5
BiPointNet	EMA	1	1	86.1
Proposed method (Hist)	APSS1^*	1	1	86.5
	APSS2^*	1	1	85.3
Proposed method (KDE)	APSS1^*	1	1	87.5
	APSS2^*	1	1	87.3
Proposed method (KNN)	APSS1^*	1	1	86.6
	APSS2^*	1	1	87.4

Table 5. Precision of part segmentation%

View table

Table 5. Precision of part segmentation%

Method	Aero	Bag	Ca	Car	Chair	Earphone	Guitar	Knife	Lamp	Laptop	Motor-bike	Mug	Pistol	Rocket	Skateboard	Table
Fullprecision	83.1	89.0	95.2	78.3	90.4	78.1	93.3	92.9	81.9	97.9	70.7	95.9	81.6	57.4	74.8	81.5
BiPointNet	79.6	69.6	86.3	67.5	88.6	69.8	87.5	83.3	75.0	95.3	45.1	91.6	76.8	47.9	57.5	79.6
Proposedmodel	80.2	64.8	87.0	66.8	87.0	77.6	89.7	84.3	76.3	96.7	50.2	92.3	79.6	50.1	66.2	80.1

Table 6. Semantic segmentation experiment results%

View table

Table 6. Semantic segmentation experiment results%

Method	OverallmIoU		Overall acc			mIoU/accof area1		mIoU/accof area2		mIoU/accof area3			mIoU/accof area4		mIoU/accof area5		mIoU/accof area6
Fullprecision	51.9		82.0			59.7/85.1		34.7/73.6		60.9/87.2			43.6/80.9		43.1/82.0		66.2/88.1
BiPointnet	43.4		76.3			50.1/77.9		29.7/69.8		53.3/81.6			36.2/73.3		36.5/77.0		57.8/82.4
Proposedmodel	43.9		77.5			51.8/78.9		27.1/68.3		55.1/83.2			37.5/75.1		36.9/78.8		59.1/84.0
Method	IoU ofceiling	IoU offloor		IoU ofwall	IoU ofbeam		IoU ofcolumn	IoU ofwindow	IoU ofdoor		IoU oftable	IoU ofchair		IoU ofsofa	IoU ofbookcase	IoU ofboard		IoU ofclutter
Fullprecision	89.7	93.7		71.0	50.2		34.0	52.9	53.4		56.7	46.6		9.5	38.5	36.4		41.3
BiPointNet	84.2	85.6		62.0	32.8		22.9	41.7	47.3		45.2	39.5		9.1	35.3	25.8		33.2
Proposedmodel	85.0	86.3		60.3	33.7		24.2	43.4	46.5		46.6	41.1		8.7	34.2	26.5		34.7

Table 7. Comparative experiment results for typical network models

View table

Table 7. Comparative experiment results for typical network models

Methods		Bit width N_w /bit	Bit width N_a /bit	Precision P_c /%
	Full precision	32	32	88.2
PointNet	BiPointNet	1	1	86.1
	Proposed method	1	1	87.2
	Full precision	32	32	90.7
PointNet++	BiPointNet	1	1	88.5
	Proposed method	1	1	89.0
	Full precision	32	32	89.7
PointCNN	BiPointNet	1	1	81.5
	Proposed method	1	1	82.8
	Full precision	32	32	90.9
DGCNN	BiPointNet	1	1	75.0
	Proposed method	1	1	83.6

Table 8. Complexity comparison results

View table

Table 8. Complexity comparison results

Method	Pooling type	FLOP persample /Mbit	Speedup ratioS_r	Parameter P_a /Mbit	Compressionratio C_r
Full precision	MAX	443.38	1	3.48	1
	MAX	8.35	53	0.15	23
BNN	APSS1^*	10.45	42	0.15	23
	APSS2^*	12.56	35	0.15	23
	MAX	8.94	50	0.16	22
IRNet	APSS1^*	11.05	40	0.16	22
	APSS2^*	13.15	34	0.16	22
	MAX	9.89	45	0.62	6
XNOR-Net	APSS1^*	11.99	37	0.62	6
	APSS2^*	14.09	31	0.62	6
	EMA	10.56	42	0.15	23
BiPointNet	APSS1^*	10.56	42	0.15	23
	APSS2^*	12.66	35	0.15	23
	MAX	8.46	52	0.15	23
Proposed model	APSS1^*	10.56	42	0.15	23
	APSS2^*	12.66	35	0.15	23

Tools

Get Citation

Copy Citation Text

Zhi Zhao, Yanxin Ma, Ke Xu, Jianwei Wan. Optimized Scalable and Learnable Binary Quantization Network for LiDAR Point Cloud[J]. Acta Optica Sinica, 2022, 42(12): 1212005

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Instrumentation, Measurement and Metrology

Received: Dec. 29, 2021

Accepted: Mar. 25, 2022

Published Online: Jun. 15, 2022

The Author Email: Zhao Zhi (zhaozhi@nudt.edu.cn)

DOI:10.3788/AOS202242.1212005

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology