It doesn't sound like you want FPKM/RPKM exactly. Those metrics normalize for gene length and also library depth so that (in theory) you can compare transcript expression levels between genes and libraries. This doesn't make sense if you want to understand how well your exome-seq experiment has worked. If one library has twice as many reads and gives you twice as good coverage you want to know that, not try to normalize that difference away, correct? You probably want average depth of coverage for a particular exon/gene or averaged across exons/genes. Search the forum for "exon coverage calculation" and you will find many posts on this topic.

• Calculate per exon/per gene coverage
• calculate exon coverage using SAMtools
• tools to calculate average coverage for a bam file?
• Calculate coverage from BAM file
• How to calculate the coverage of a specific interval based on data from a bam file
• NGS and coverage of exons

Try using qualimap it should give you coverage information.

You don't mean FPKM or RPKM. You want to use GATK's depthofcoverage walker. You can give it as inputs your bam files and your target, and get depth of coverage across your entire target, or if you just want depth of coverage of a certain gene or genes you can use the -L option to input a file with those specific genes. Take a look at GATK's documentation for the correct formats of the target input files.