Recognizing Human Action at a Distance in Video by Key Poses
Citations
294 citations
Cites methods from "Recognizing Human Action at a Dista..."
...The approaches developed based on video sequences can be classified into template-based approaches, where emphasis is placed on low- and mid-level features, and model-based approaches where emphasis is placed on high-level features [45]....
[...]
152 citations
Cites methods from "Recognizing Human Action at a Dista..."
...The methods of human action recognition from image frames or video sequences are broadly classified as templatebased approach (emphasis on collecting low- and mid-level features) and model-based approach (emphasis on feature for high-level interaction) [7]....
[...]
76 citations
Cites methods from "Recognizing Human Action at a Dista..."
...HoF features were calculated according to [58] in a hierarchical way using a pyramid....
[...]
49 citations
Cites background from "Recognizing Human Action at a Dista..."
...It is captured in a controlled environment with simple background and holds camera motion and zooming effect in few videos (Mukherjee et al., 2011)....
[...]
39 citations
Cites background or methods from "Recognizing Human Action at a Dista..."
...We use motion and oriented gradient information to build a pose descriptor of each frame [21]....
[...]
...For the details, please go through [21]....
[...]
...Where, in [21] they starts with a large vocabulary of poses (visual words) and derives a refined and bharatnatyam 83....
[...]
...Moreover, it is significantly better than another related algorithm proposed in [21]....
[...]
References
31,952 citations
"Recognizing Human Action at a Dista..." refers background or result in this paper
...Table I shows that hierarchical implementations of HOOF/HOG in multiple layers work better than their raw counterparts, but our proposed hierarchical HOOF computed on gradient weighted optic flow [using (1)] outperforms all others....
[...]
...We show in Table I the results of our experiments using raw HOOF [20] (row 3) and HOG [21] (row 4) features as well as hierarchical HOOF without using (1) (row 5) and hierarchical HOG (row 6)....
[...]
...HOG feature on 3-layer architecture [21] 98 74....
[...]
...HOG feature on 3-layer architecture [21] 62....
[...]
...2(a)] and the histogram of oriented gradient (HOG) [21] [derived from Fig....
[...]
22,840 citations
18,802 citations
"Recognizing Human Action at a Dista..." refers methods in this paper
...An optimum (local) lower bound on the codebook size of S can be estimated by Akaike information criterion, or Bayesian information criterion [23] or one can directly employ X-means algorithm [24]....
[...]
17,104 citations