What are some conferences for publishing papers on Deep Learning for Human Activity recognition? Do any of the major conferences have specific tracks for Human Activity Recognition?
Under the assumption, that the papers at Google Scholar are expressing the current state of technology it's easy to take a closer look into the list to determine in which conference proceedings they were published in the past. The first information is, that a dedicated conference on human activity recognition isn't available, but the topic is spread over different conferences which have to do with Pervasive computing, mobile computing, image analysts and wearable systems.
The hypothesis is, that the topic isn't important enough to create a dedicated conference but instead, it's a subtopic in existing AI-related proceedings. Another option is, that the topic is perceived as important but the academic community hasn't figured out how to solve the issue. The problem is, that human activity recognition consists of many sub-disciplines which are on the lower level the pixel recognition and on the higher level a linguistic understanding of longer sequences. The EU founded Poeticon++ project was the first attempt to realize an integrated architecture, but this was not a conference with different participants but a single workgroup who have ended after a while.
If your using computer vision, the top recognised conference is CVPR (computer vision and pattern recognition) . You can also submit at ICML(International conference on machine learning) and NIPS(Neural Information Processing Systems) which focuses on applications of machine learning and deep learning. I'd also recommend IJCAI (International joint conference on artificial intelligence).