As a video editor, this astonishes me. For VFX shots, you are compositing these 3D models on top of the captured footage. This means that whenever the vehicles are viewed through a chain-link fence or are passed by a bus, the animator took the shots frame by frame to add in the models to appear below the fence. Additionally, the handheld feel (probably added in post) sells the entire piece. Fantastic job by a talented artist!