ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Incremental Learning for Animal Pose Estimation using RBF k-DPP

Nayak, GK and Shah, H and Chakraborty, A (2021) Incremental Learning for Animal Pose Estimation using RBF k-DPP. In: UNSPECIFIED.

[img] PDF
32_Bri_mac_vis_cn_2021.pdf - Published Version
Restricted to Registered users only

Download (2MB) | Request a copy
Official URL: https://www.scopus.com/record/display.uri?eid=2-s2...


Pose estimation is the task of locating keypoints for an object of interest in an image. Animal Pose estimation is more challenging than estimating human pose due to high inter and intra class variability in animals. Existing works solve this problem for a fixed set of predefined animal categories. Models trained on such sets usually do not work well with new animal categories. Retraining the model on new categories makes the model overfit and leads to catastrophic forgetting. Thus, in this work, we propose a novel problem of �Incremental Learning for Animal Pose Estimation�. Our method uses an exemplar memory, sampled using Determinantal Point Processes (DPP) to continually adapt to new animal categories without forgetting the old ones. We further propose a new variant of k-DPP that uses RBF kernel (termed as �RBF k-DPP�) which gives more gain in performance over traditional k-DPP. Due to memory constraints, the limited number of exemplars along with new class data can lead to class imbalance. We mitigate it by performing image warping as an augmentation technique. This helps in crafting diverse poses, which reduces overfitting and yields further improvement in performance. The efficacy of our proposed approach is demonstrated via extensive experiments and ablations where we obtain significant improvements over state-of-the-art baseline methods. © 2021. The copyright of this document resides with its authors.

Item Type: Conference Paper
Publication: 32nd British Machine Vision Conference, BMVC 2021
Publisher: British Machine Vision Association, BMVA
Additional Information: The copyright for this article belongs to British Machine Vision Association, BMVA.
Keywords: Computer vision; Radial basis function networks, Catastrophic forgetting; Fixed sets; Human pose; Incremental learning; Inter class; Intra class; Keypoints; Performance; Point process; Pose-estimation, Animals
Department/Centre: Division of Biological Sciences > Molecular Reproduction, Development & Genetics
Division of Biological Sciences > Central Animal Facility (Formerly Primate Research Laboratory)
Date Deposited: 04 Mar 2024 09:52
Last Modified: 04 Mar 2024 09:52
URI: https://eprints.iisc.ac.in/id/eprint/84415

Actions (login required)

View Item View Item