Coupling Top-down and Bottom-up Methods for 3D Human Pose and Shape Estimation from Monocular Image Sequences
This work addresses the need for inferring human bio-signatures (e.g., size, shape) from surveillance videos, which is important for intelligence and security applications, but it appears incremental as it builds on existing discriminative and tracking methods.
The paper tackles the ill-posed problem of estimating 3D human pose and shape from monocular image sequences, which is challenging due to ambiguities like self-occlusion. It presents a system that integrates learned regression models and likelihood feedback to achieve robust, automated tracking and shape fitting, evaluated on a large dataset.
Until recently Intelligence, Surveillance, and Reconnaissance (ISR) focused on acquiring behavioral information of the targets and their activities. Continuous evolution of intelligence being gathered of the human centric activities has put increased focus on the humans, especially inferring their innate characteristics - size, shapes and physiology. These bio-signatures extracted from the surveillance sensors can be used to deduce age, ethnicity, gender and actions, and further characterize human actions in unseen scenarios. However, recovery of pose and shape of humans in such monocular videos is inherently an ill-posed problem, marked by frequent depth and view based ambiguities due to self-occlusion, foreshortening and misalignment. The likelihood function often yields a highly multimodal posterior that is difficult to propagate even using the most advanced particle filtering(PF) algorithms. Motivated by the recent success of the discriminative approaches to efficiently predict 3D poses directly from the 2D images, we present several principled approaches to integrate predictive cues using learned regression models to sustain multimodality of the posterior during tracking. Additionally, these learned priors can be actively adapted to the test data using a likelihood based feedback mechanism. Estimated 3D poses are then used to fit 3D human shape model to each frame independently for inferring anthropometric bio-signatures. The proposed system is fully automated, robust to noisy test data and has ability to swiftly recover from tracking failures even after confronting with significant errors. We evaluate the system on a large number of monocular human motion sequences.