1adrianb / 1adrianb/binary-human-pose-estimation

how to extract multi-scale feature by using Fig.4e block?

Aperta
#16 0 commenti 0 reazioni 0 assegnatari Vedi su GitHub
Lingua principale
Lua
Stelle
214
Fork
67
Metriche di merge delle PR
Nessuna PR unita negli ultimi 30g

Descrizione

hi,
I have a question about the final block.
In your paper, you mentioned multi-scale should be used in Fig.4e which is your final block. However, Fig.4e only contains 3x3 conv. It doesn't show any pooling or conv with stride operation. I can't understand how to extract multi-scale feature by using it. Could you explain more details about it? Thanks a lot.
![image](https://user-images.githubusercontent.com/47654539/60490432-1a9f6200-9cd9-11e9-9e31-1cd2f26829e4.png)

Guida per i contributori

Nessuna guida per i contributori indicizzata per questo repository

Direzione di ricerca

The issue references Figure 4e from the paper and asks about multi-scale feature extraction. The repository contains Lua/Torch7 code for human pose estimation. To understand the implementation, examine the model architecture files (likely in the root or a models/ directory) for the final block. Look for convolutional layers, pooling, or stride operations. Compare the code to the paper's description. Determine if the figure is accurately implemented or if there's a discrepancy. Running the demo might help visualize the feature maps.

Scritto dal modello di indicizzazione a partire dal testo della issue.

Valutazione

Stack tecnologico
computer-vision, lua
Ambito
computer-vision, machine-learning
Tipo di issue
Documentazione
Difficoltà
4/5
Tempo stimato
3-5 giorni
Stato di attività
Ferma
Chiarezza
Da chiarire
Idoneità per principianti
35/100

Ricevi le nuove issue nella tua casella

Un breve riepilogo di issue GitHub adatte ai principianti.