Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for estimating a 3-D pose of an object of interest from image and point cloud data. In one aspect, a method includes obtaining an image of an environment; obtaining a point cloud of a three-dimensional region of the environment; generating a fused representation of the image and the point cloud; and processing the fused representation using a pose estimation neural network and in accordance with current values of a plurality of pose estimation network parameters to generate a pose estimation network output that specifies, for each of multiple keypoints, a respective estimated position in the three-dimensional region of the environment.

Abstract: A combined sofa in a steel structure includes a sofa base, a backrest connected with the sofa base and a sofa armrest arranged on at least one end of the sofa base, wherein, the sofa base comprises a seat base frame, on which spring assembly is laid. The backrest is detachably connected to the rear side of the sofa base. With this design, during transportation, the backrest and armrest of the sofa can be disassembled, and the sofa base can be disassembled into one or a plurality of independent seat base frames, which is very convenient for transportation. The seat base frame(s) can be combined again as required for use to realize variable structures.

Abstract: Methods and apparatuses are disclosed that automatically generate relative placement rules. Constructs at the register transfer language-level result in relative placement rules specified at the register transfer language-level.

Abstract: Methods and apparatuses are disclosed that automatically generate relative placement rules. Constructs at the register transfer language-level result in relative placement rules specified at the register transfer language-level.