Hi,
Im currently studying Graphic design and would appreciate some feedback on my work. These are just some practice pieces to help learn some new techniques before I proceed to the final project. I just want to understand the best way to blend images seamlessly. Any thoughts or help would be greatly appreciated. 
Your blending of the solar system around the scoop and the smoke head looks really good. The one area that I see could use some work is the lighting on the top scoop. Think about where your light source use. There is a strong highlight across the bottom of the top scoop that doesn’t seem natural.
I swear I see the bottom of the arm going into the floor. It might not be but, be aware of anything that can throw the eye off from what ever elements you are using.
I really like the last one. The smoke coming out of the shirt looks great. Just be really careful with the white outline that can be left behind on dark colors, such as with this coat.
Other than that and what Steve mentioned, you are coming along nicely 
Yeah, there is some transparent forearm below the violet frame. Plus, the perspective of the frame is mismatched to that of the floor.
1 Like
The last one is really awesome - smoke coming out from the shirt.
If the smoke is also coming out of the cup, then it adds beauty…
The hand in the blue/purple neon composition looks a bit cutout. You can soften the edges to make it look less cutout. It seems that you have the seamless part accurate on the other 2 photos. It’s the lighting and shadow effects that show plenty of room for improvement. But don’t be discouraged. From my 25 years experience, one can never do enough when it comes to lighting and shadow effects. If you are critical and honest with yourself, you will always find something that could use improving with your own work. What matters is whether someone else can find it.
With surrealistic and hyperrealistic compositions, the ultimate compliment is when a person can’t tell whether the photo effect is a special effect (SFX, made from a prop) or a visual effect (VFX, digital editing) made post production. For example, if someone thinks that you pumped real colored smoke out of a headless dummy, you know that there’s nothing more you can do to make it look any more real. Likewise, if someone thinks that that wooden background was added to the photo with the bench and person separately, then you know you aren’t dealing with a sophisticated photo manipulation critic, and should not take their opinion too seriously.
When it comes to lighting and shadowing VFX, 3D modeling blows away 2D photo editing. 3D modeling is using virtual props and a virtual camera lens. Sometimes it’s less time-consuming. I’ve done projects when I’ve modeled the scene using 3D figures just to get the shadows and lighting right, and then replaced the figures with real photos. When things don’t look quite right, it’s a lot easier to tweak the model and re-render than it is to tweak a 2D photo. 2D photos often get an over-edited look when tweaked.