Individual recognition using gait energy image
Citations
984 citations
Cites background from "Individual recognition using gait e..."
...Its features include HOG3D [106] and the gait energy image (GEI) [107]....
[...]
900 citations
Cites methods from "Individual recognition using gait e..."
...In our experiment, the size of GEI is 80× 30, which is reshaped into a column as the final vector....
[...]
...4 and Table 4, we observe that HOG3D and GEI feature both yield decent accuracy on iLIDS-VID and PRID-2011 datasets....
[...]
...The Gait Energy Image (GEI) [13] is widely applied in gait recognition....
[...]
...Performance of HOG3D and GEI....
[...]
...The resulting binary images within a cycle are averaged to yield the GEI of the tracklet....
[...]
862 citations
856 citations
Cites methods from "Individual recognition using gait e..."
...recommended by the authors of [5], and the ETGLDA results shown are obtained with , determined empirically....
[...]
600 citations
Cites background or methods from "Individual recognition using gait e..."
...It is also assumed that complete gait cycles were captured in target image sequences [17,31]....
[...]
...Gait Recognition - Space-time information of sequences has been extensively exploited by gait recognition [33,38,17,31]....
[...]
...We compared the proposed DVR model with contemporary gait recognition and temporal sequence matching methods for person (re-)identification: (1) Gait recognition (GEI+RSVM) [31]: A state-of-the-art gait recognition model using Gait Energy Image (GEI) [17] (computed from pre-segmented silhouettes in their datasets) as sequence representation and RankSVM [6] for recognition....
[...]
...Comparison with Gait Recognition and Temporal Sequence Matching - We compared the proposed DVR model with contemporary gait recognition and temporal sequence matching methods for person (re-)identification: (1) Gait recognition (GEI+RSVM) [31]: A state-of-the-art gait recognition model using Gait Energy Image (GEI) [17] (computed from pre-segmented silhouettes in their datasets) as sequence representation and RankSVM [6] for recognition....
[...]
...This is largely due to very noisy GEI features avaliable from person sequences....
[...]
References
20,541 citations
"Individual recognition using gait e..." refers background in this paper
...As described in [6], PCA seeks a projection that best represents the data in a least square sense, while MDA seeks a projection that best separates the data in a least-square sense....
[...]
11,674 citations
2,932 citations
"Individual recognition using gait e..." refers background in this paper
...Unlike other gait representations [8, 4] which consider gait as a sequence of templates (poses), GEI represents human motion sequence in a single image while preserving some temporal information....
[...]
...Bobick and Davis [4] propose motion-energy image (MEI) and motion-history image (MHI) for human movement recognition....
[...]
2,428 citations
Additional excerpts
...There are two classical linear approaches for finding transformations for dimensionality reduction—Principal Component Analysis (PCA) and Multiple Discriminant Analysis (MDA) that have been effectively used in face recognition [12]....
[...]
1,157 citations
"Individual recognition using gait e..." refers background or methods in this paper
...A silhouette preprocessing procedure [5] is then applied on the extracted silhouette sequences....
[...]
...(Legends: USF—direct frame shape matching [5]; CMU—key frame shape matching [6]; UMD—HMM framework [3]; real—proposed real gait feature classifier only; sync—proposed synthetic gait feature classifier only; fusion—proposed gait feature fusion....
[...]
...Table 2 compares the recognition performance of USF baseline algorithm [5] and our proposed approach....
[...]
...While some gait recognition approaches [4] extract features from the correlation of all the frames in a walking sequence without considering their order, other approaches extract features from each frame and compose a feature sequence for the human walking sequence [2], [5], [6]....
[...]
...(Legends: baseline—direct frame shape matching [5]; real—proposed real gait feature classifier only; synthetic—proposed synthetic gait feature classifier only; fusion—proposed gait feature fusion....
[...]