公开(公告)号：US12154188B2

公开(公告)日：2024-11-26

申请号：US17890849

申请日：2022-08-18

Applicant: NVIDIA Corporation

Inventor： Fnu Ratnesh Kumar ,  Farzin Aghdasi ,  Parthasarathy Sriram ,  Edwin Weill

IPC: G06T1/20 ,  G06F17/18 ,  G06N3/045 ,  G06N3/047 ,  G06N3/08 ,  G06V10/764 ,  G06V10/82 ,  G06V20/52 ,  G06V20/58

Abstract: In various examples, a neural network may be trained for use in vehicle re-identification tasks—e.g., matching appearances and classifications of vehicles across frames—in a camera network. The neural network may be trained to learn an embedding space such that embeddings corresponding to vehicles of the same identify are projected closer to one another within the embedding space, as compared to vehicles representing different identities. To accurately and efficiently learn the embedding space, the neural network may be trained using a contrastive loss function or a triplet loss function. In addition, to further improve accuracy and efficiency, a sampling technique—referred to herein as batch sample—may be used to identify embeddings, during training, that are most meaningful for updating parameters of the neural network.

公开(公告)号：US20250022092A1

公开(公告)日：2025-01-16

申请号：US18825755

申请日：2024-09-05

Applicant: NVIDIA Corporation

Inventor： Fnu Ratnesh Kumar ,  Farzin Aghdasi ,  Parthasarathy Sriram ,  Edwin Weill

IPC: G06T1/20 ,  G06F17/18 ,  G06N3/045 ,  G06N3/047 ,  G06N3/08 ,  G06V10/764 ,  G06V10/82 ,  G06V20/52 ,  G06V20/58

Abstract: In various examples, a neural network may be trained for use in vehicle re-identification tasks—e.g., matching appearances and classifications of vehicles across frames—in a camera network. The neural network may be trained to learn an embedding space such that embeddings corresponding to vehicles of the same identify are projected closer to one another within the embedding space, as compared to vehicles representing different identities. To accurately and efficiently learn the embedding space, the neural network may be trained using a contrastive loss function or a triplet loss function. In addition, to further improve accuracy and efficiency, a sampling technique—referred to herein as batch sample—may be used to identify embeddings, during training, that are most meaningful for updating parameters of the neural network.

公开(公告)号：US11455807B2

公开(公告)日：2022-09-27

申请号：US16577716

申请日：2019-09-20

Applicant: NVIDIA Corporation

Inventor： Fnu Ratnesh Kumar ,  Farzin Aghdasi ,  Parthasarathy Sriram ,  Edwin Weill

IPC: G06K9/00 ,  G06V20/58 ,  G06F17/18 ,  G06N3/04 ,  G06N3/08 ,  G06T1/20

Abstract: In various examples, a neural network may be trained for use in vehicle re-identification tasks—e.g., matching appearances and classifications of vehicles across frames—in a camera network. The neural network may be trained to learn an embedding space such that embeddings corresponding to vehicles of the same identify are projected closer to one another within the embedding space, as compared to vehicles representing different identities. To accurately and efficiently learn the embedding space, the neural network may be trained using a contrastive loss function or a triplet loss function. In addition, to further improve accuracy and efficiency, a sampling technique—referred to herein as batch sample—may be used to identify embeddings, during training, that are most meaningful for updating parameters of the neural network.

公开(公告)号：US20220392234A1

公开(公告)日：2022-12-08

申请号：US17890849

申请日：2022-08-18

Applicant: NVIDIA Corporation

Inventor： Fnu Ratnesh Kumar ,  Farzin Aghdasi ,  Parthasarathy Sriram ,  Edwin Weill

IPC: G06V20/58 ,  G06F17/18 ,  G06N3/04 ,  G06N3/08 ,  G06T1/20

Abstract: In various examples, a neural network may be trained for use in vehicle re-identification tasks—e.g., matching appearances and classifications of vehicles across frames—in a camera network. The neural network may be trained to learn an embedding space such that embeddings corresponding to vehicles of the same identify are projected closer to one another within the embedding space, as compared to vehicles representing different identities. To accurately and efficiently learn the embedding space, the neural network may be trained using a contrastive loss function or a triplet loss function. In addition, to further improve accuracy and efficiency, a sampling technique—referred to herein as batch sample—may be used to identify embeddings, during training, that are most meaningful for updating parameters of the neural network.

公开(公告)号：US20200097742A1

公开(公告)日：2020-03-26

申请号：US16577716

申请日：2019-09-20

Applicant: NVIDIA Corporation

Inventor： Fnu Ratnesh Kumar ,  Farzin Aghdasi ,  Parthasarathy Sriram ,  Edwin Weill

IPC: G06K9/00 ,  G06N3/04 ,  G06N3/08 ,  G06F17/18 ,  G06T1/20

Abstract: In various examples, a neural network may be trained for use in vehicle re-identification tasks—e.g., matching appearances and classifications of vehicles across frames—in a camera network. The neural network may be trained to learn an embedding space such that embeddings corresponding to vehicles of the same identify are projected closer to one another within the embedding space, as compared to vehicles representing different identities. To accurately and efficiently learn the embedding space, the neural network may be trained using a contrastive loss function or a triplet loss function. In addition, to further improve accuracy and efficiency, a sampling technique—referred to herein as batch sample—may be used to identify embeddings, during training, that are most meaningful for updating parameters of the neural network.