Long term Recurrent Convolutional Networks for Visual Recognition and Descriptio

Long-term Recurrent Convolutional Networks for Visual Recognition and DescriptionJeff Donahue? Lisa Anne Hendricks? Sergio Guadarrama? Marcus Rohrbach?∗Subhashini Venugopalan† †UT Austin Austin, TXvsub@cs.utexas.eduKate Saenko‡ ‡UMass LowellLowell, MA saenko@cs.uml.eduTrevor Darrell?∗ ?UC Berkeley, ∗ICSIBerkeley, CA {jdonahue, lisa anne, sguada,rohrbach, trevor}@eecs.berkeley.eduAbstractModels based on deep convolutional networks have dom- inated recent image interpretation tasks; we investigate