Picture for Michael Greenspan

Michael Greenspan

Diffusion Dataset Generation: Towards Closing the Sim2Real Gap for Pedestrian Detection

May 16, 2023
Andrew Farley, Mohsen Zand, Michael Greenspan

* 8 pages, 4 figures, Accepted to CRV2023 conference

Access Paper or Ask Questions

ObjectBox: From Centers to Boxes for Anchor-Free Object Detection

Jul 14, 2022
Mohsen Zand, Ali Etemad, Michael Greenspan

Figure 1 for ObjectBox: From Centers to Boxes for Anchor-Free Object Detection

Figure 2 for ObjectBox: From Centers to Boxes for Anchor-Free Object Detection

Figure 3 for ObjectBox: From Centers to Boxes for Anchor-Free Object Detection

* ECCV 2022 Oral

Access Paper or Ask Questions

Neuro-symbolic Natural Logic with Introspective Revision for Natural Language Inference

Mar 09, 2022
Yufei Feng, Xiaoyu Yang, Xiaodan Zhu, Michael Greenspan

Figure 1 for Neuro-symbolic Natural Logic with Introspective Revision for Natural Language Inference

Figure 2 for Neuro-symbolic Natural Logic with Introspective Revision for Natural Language Inference

Figure 3 for Neuro-symbolic Natural Logic with Introspective Revision for Natural Language Inference

Figure 4 for Neuro-symbolic Natural Logic with Introspective Revision for Natural Language Inference

* To appear at TACL 2022, MIT Press

Access Paper or Ask Questions

Multiscale Crowd Counting and Localization By Multitask Point Supervision

Feb 21, 2022
Mohsen Zand, Haleh Damirchi, Andrew Farley, Mahdiyar Molahasani, Michael Greenspan, Ali Etemad

Figure 1 for Multiscale Crowd Counting and Localization By Multitask Point Supervision

Figure 2 for Multiscale Crowd Counting and Localization By Multitask Point Supervision

Figure 3 for Multiscale Crowd Counting and Localization By Multitask Point Supervision

Figure 4 for Multiscale Crowd Counting and Localization By Multitask Point Supervision

* 4 pages + references, 3 figures, 2 tables, Accepted by ICASSP 2022 Conference

Access Paper or Ask Questions

Multistream ValidNet: Improving 6D Object Pose Estimation by Automatic Multistream Validation

Jun 12, 2021
Joy Mazumder, Mohsen Zand, Michael Greenspan

Figure 1 for Multistream ValidNet: Improving 6D Object Pose Estimation by Automatic Multistream Validation

Figure 2 for Multistream ValidNet: Improving 6D Object Pose Estimation by Automatic Multistream Validation

Figure 3 for Multistream ValidNet: Improving 6D Object Pose Estimation by Automatic Multistream Validation

Figure 4 for Multistream ValidNet: Improving 6D Object Pose Estimation by Automatic Multistream Validation

* 6 pages, 2 figures, 2 tables. To appear in the proceedings of the 28th IEEE International Conference on Image Processing (IEEE - ICIP), September 19-22, 2021, Anchorage, Alaska, USA

Access Paper or Ask Questions

Oriented Bounding Boxes for Small and Freely Rotated Objects

Apr 24, 2021
Mohsen Zand, Ali Etemad, Michael Greenspan

Figure 1 for Oriented Bounding Boxes for Small and Freely Rotated Objects

Figure 2 for Oriented Bounding Boxes for Small and Freely Rotated Objects

Figure 3 for Oriented Bounding Boxes for Small and Freely Rotated Objects

Figure 4 for Oriented Bounding Boxes for Small and Freely Rotated Objects

* IEEE Transactions on Geoscience and Remote Sensing, 2021

Access Paper or Ask Questions

Teacher-Student Adversarial Depth Hallucination to Improve Face Recognition

Apr 06, 2021
Hardik Uppal, Alireza Sepas-Moghaddam, Michael Greenspan, Ali Etemad

Figure 1 for Teacher-Student Adversarial Depth Hallucination to Improve Face Recognition

Figure 2 for Teacher-Student Adversarial Depth Hallucination to Improve Face Recognition

Figure 3 for Teacher-Student Adversarial Depth Hallucination to Improve Face Recognition

Figure 4 for Teacher-Student Adversarial Depth Hallucination to Improve Face Recognition

* 10 pages, 6 figures

Access Paper or Ask Questions