High-resolution spectra in the near-infrared (NIR) are an important tool for the detailed study of stellar atmospheres. The accurate identification of elements and molecules in these spectra can be used to determine chemical abundances and physical conditions in the photosphere of the observed star. Such identifications require precise line positions and strengths of both atomic and molecular features. This work focuses on the full identification of absorption lines in the NIR spectrum of the K giant 10 Leo, including previously unidentified lines. The large number and complexity of the observed absorption lines require a deep search for potential spectral signatures, to enable unambiguous assignment to specific elements or molecular species. We aim at improving published line lists of metals, some of which are determined by model calculations only, and many of which presently lack completeness and accuracy of line parameters. The CRIRES-POP (CRyogenic high-resolution InfraRed Echelle Spectrograph-Paranal Observatory Project) provided high-resolution, high signal-to-noise (S/N) ratio spectra of several bright stars in the 1 to 5um range. For the K-giant 10 Leo, a spectrum corrected for telluric absorption and with precise wavelength calibration is available. This has been analysed by comparison with model spectra and up-to-date line lists. We identified lines of 29 elements and 8 molecular species. While the positions of many known lines could be confirmed, about 6% of all lines detected in 10 Leo could not be attributed to any known feature. For CO and its isotopologues, molecular constants could be derived and several additional lines identified. We report major inconsistencies for some prominent lines. In addition, abundances for several key elements in 10 Leo are provided.