Benchmark Results - ScanRefer Benchmark

This table lists the benchmark results for the ScanRefer Localization Benchmark scenario.

		Unique	Unique	Multiple	Multiple	Overall	Overall
Method	Info	acc@0.25IoU	acc@0.5IoU	acc@0.25IoU	acc@0.5IoU	acc@0.25IoU	acc@0.5IoU
Method	Info
ConcreteNet		0.8120 3	0.6933 1	0.4479 4	0.3760 2	0.5296 3	0.4471 1

M3DRef-CLIP		0.8170 1	0.6925 2	0.4582 2	0.3617 3	0.5386 1	0.4359 2

HAM		0.7799 8	0.6373 10	0.4148 10	0.3324 4	0.4967 9	0.4007 3
Jiaming Chen, Weixin Luo, Xiaolin Wei, Lin Ma, Wei Zhang: HAM: Hierarchical Attention Model with High Performance for 3D Visual Grounding.
CSA-M3LM		0.8137 2	0.6241 11	0.4544 3	0.3317 5	0.5349 2	0.3972 4

D3Net		0.7923 4	0.6843 3	0.3905 14	0.3074 8	0.4806 12	0.3919 5
Dave Zhenyu Chen, Qirui Wu, Matthias Niessner, Angel X. Chang: D3Net: A Unified Speaker-Listener Architecture for 3D Dense Captioning and Visual Grounding. 17th European Conference on Computer Vision (ECCV), 2022
ContraRefer		0.7832 7	0.6801 6	0.3850 15	0.2947 9	0.4743 13	0.3811 6

Clip		0.7733 12	0.6810 5	0.3619 18	0.2919 13	0.4542 17	0.3791 7

Clip-pre		0.7766 10	0.6843 3	0.3617 20	0.2904 14	0.4547 16	0.3787 8

3DJCG(Grounding)		0.7675 15	0.6059 12	0.4389 5	0.3117 6	0.5126 4	0.3776 9
Daigang Cai, Lichen Zhao, Jing Zhang†, Lu Sheng, Dong Xu: 3DJCG: A Unified Framework for Joint Dense Captioning and Visual Grounding on 3D Point Clouds. CVPR2022 Oral
3DVG-Trans +		0.7733 12	0.5787 17	0.4370 6	0.3102 7	0.5124 5	0.3704 10
Lichen Zhao∗, Daigang Cai∗, Lu Sheng†, Dong Xu: 3DVG-Transformer: Relation Modeling for Visual Grounding on Point Clouds. ICCV2021
FE-3DGQA		0.7857 5	0.5862 16	0.4317 7	0.2935 10	0.5111 6	0.3592 11

D3Net - Pretrained		0.7659 16	0.6579 8	0.3619 18	0.2726 16	0.4525 19	0.3590 12
Dave Zhenyu Chen, Qirui Wu, Matthias Niessner, Angel X. Chang: D3Net: A Unified Speaker-Listener Architecture for 3D Dense Captioning and Visual Grounding. 17th European Conference on Computer Vision (ECCV), 2022
HGT		0.7692 14	0.5886 15	0.4141 11	0.2924 12	0.4937 10	0.3588 13

InstanceRefer		0.7782 9	0.6669 7	0.3457 22	0.2688 17	0.4427 21	0.3580 14
Zhihao Yuan, Xu Yan, Yinghong Liao, Ruimao Zhang, Zhen Li*, Shuguang Cui: InstanceRefer: Cooperative Holistic Understanding for Visual Grounding on Point Clouds through Instance Multi-level Contextual Referring. ICCV 2021
3DVG-Transformer		0.7576 17	0.5515 19	0.4224 9	0.2933 11	0.4976 8	0.3512 15
Lichen Zhao∗, Daigang Cai∗, Lu Sheng†, Dong Xu: 3DVG-Transformer: Relation Modeling for Visual Grounding on Point Clouds. ICCV2021
SAVG		0.7758 11	0.5664 18	0.4236 8	0.2826 15	0.5026 7	0.3462 16

PointGroup_MCAN		0.7510 18	0.6397 9	0.3271 24	0.2535 19	0.4222 23	0.3401 17

TransformerVG		0.7502 19	0.5977 13	0.3712 16	0.2628 18	0.4562 15	0.3379 18

TGNN		0.6834 24	0.5894 14	0.3312 23	0.2526 20	0.4102 24	0.3281 19
Pin-Hao Huang, Han-Hung Lee, Hwann-Tzong Chen, Tyng-Luh Liu: Text-Guided Graph Neural Network for Referring 3D Instance Segmentation. AAAI 2021
BEAUTY-DETR		0.7848 6	0.5499 20	0.3934 13	0.2480 21	0.4811 11	0.3157 20
Ayush Jain, Nikolaos Gkanatsios, Ishita Mediratta, Katerina Fragkiadaki: Looking Outside the Box to Ground Language in 3D Scenes.
grounding		0.3822 31	0.2421 31	0.7298 1	0.5458 1	0.4538 18	0.3046 21

henet		0.7110 21	0.5180 22	0.3936 12	0.2472 22	0.4590 14	0.3030 22

SRGA		0.7494 20	0.5128 23	0.3631 17	0.2218 23	0.4497 20	0.2871 23

SR-GAB		0.7016 22	0.5202 21	0.3233 25	0.1959 26	0.4081 25	0.2686 24

SPANet		0.5614 30	0.4641 25	0.2800 29	0.2071 25	0.3431 30	0.2647 25

ScanRefer		0.6859 23	0.4353 26	0.3488 21	0.2097 24	0.4244 22	0.2603 26
Dave Zhenyu Chen, Angel X. Chang, Matthias Nießner: ScanRefer: 3D Object Localization in RGB-D Scans using Natural Language. 16th European Conference on Computer Vision (ECCV), 2020
scanrefer2		0.6340 27	0.4353 26	0.3193 26	0.1947 27	0.3898 26	0.2486 27

TransformerRefer		0.6010 28	0.4658 24	0.2540 31	0.1730 31	0.3318 31	0.2386 28

ScanRefer Baseline		0.6422 26	0.4196 28	0.3090 27	0.1832 28	0.3837 27	0.2362 29

ScanRefer_vanilla		0.6488 25	0.4056 29	0.3052 28	0.1782 29	0.3823 28	0.2292 30

pairwisemethod		0.5779 29	0.3603 30	0.2792 30	0.1746 30	0.3462 29	0.2163 31